Ontology matching consists of finding correspondences between semantically related entities of two ontologies. OAEI campaigns aim at comparing ? The official results of the campaign are on the OAEI web site.
ontology matching systems on precisely defined test cases. These test cases can use ontologies of different nature (from simple thesauri to expressive OWL ontologies) and use different modalities, e.g., blind evaluation, open evaluation, or consensus. OAEI 2016 offered 9 tracks with 22 test cases, and was attended by 21 participants. This paper is an overall presentation of the OAEI 2016 campaign.
The Ontology Alignment Evaluation Initiative1 (OAEI) is a coordinated international initiative, which organises the evaluation of an increasing number of ontology matching systems [18,21]. Its main goal is to compare systems and algorithms openly and on the same basis, in order to allow anyone to draw conclusions about the best matching strategies. Furthermore, our ambition is that, from such evaluations, tool developers can improve their systems.
Two first events were organised in 2004: (i) the Information Interpretation and
Integration Conference (I3CON) held at the NIST Performance Metrics for
Intelligent Systems (PerMIS) workshop and (ii) the Ontology Alignment Contest held at
the Evaluation of Ontology-based Tools (EON) workshop of the annual International
Semantic Web Conference (ISWC) [41]. Then, a unique OAEI campaign occurred in
2005 at the workshop on Integrating Ontologies held in conjunction with the
International Conference on Knowledge Capture (K-Cap) [
Since 2011, we have been using an environment for automatically processing evaluations (x2.2), which has been developed within the SEALS (Semantic Evaluation At Large Scale) project3. SEALS provided a software infrastructure, for automatically executing evaluations, and evaluation campaigns for typical semantic web tools, including ontology matching. In the OAEI 2016, all systems were executed under the SEALS client in all tracks, and evaluated with the SEALS client in all tracks. This year we welcomed two new tracks: the Disease and Phenotype track, sponsored by the Pistoia Alliance Ontologies Mapping project, and the Process Model Matching track. Additionally, the Instance Matching track featured a total of 7 matching tasks based on all new data sets. On the other hand, the OA4QA track was discontinued this year.
This paper synthesises the 2016 evaluation campaign. The remainder of the paper is organised as follows: in Section 2, we present the overall evaluation methodology that has been used; Sections 3-11 discuss the settings and the results of each of the test cases; Section 12 overviews lessons learned from the campaign; and finally, Section 13 concludes the paper. 2
We first present the test cases proposed this year to the OAEI participants (x2.1). Then, we discuss the resources used by participants to test their systems and the execution
environment used for running the tools (x2.2). Finally, we describe the steps of the OAEI campaign (x2.3-2.5) and report on the general execution of the campaign (x2.6). 2.1
This year’s OAEI campaign consisted of 9 tracks gathering 22 test cases, and different evaluation modalities: The benchmark track (x3): Like in previous campaigns, a systematic benchmark series has been proposed. The goal of this benchmark series is to identify the areas in which each matching algorithm is strong or weak by systematically altering an ontology. This year, we generated a new benchmark based on the original bibliographic ontology and another benchmark using a film ontology.
The expressive ontology track offers alignments between real world ontologies expressed in OWL: Anatomy (x4): The anatomy test case is about matching the Adult Mouse Anatomy (2744 classes) and a small fragment of the NCI Thesaurus (3304 classes) describing the human anatomy.
Conference (x5): The goal of the conference test case is to find all correct correspondences within a collection of ontologies describing the domain of organising conferences. Results were evaluated automatically against reference alignments and by using logical reasoning techniques.
Large biomedical ontologies (x6): The largebio test case aims at finding alignments between large and semantically rich biomedical ontologies such as FMA, SNOMED-CT, and NCI. The UMLS Metathesaurus has been used as the basis for reference alignments.
Disease & Phenotype (x7): The disease & phenotype test case aims at finding alignments between two disease ontologies (DOID and ORDO) as well as between human (HPO) and mammalian (MP) phenotype ontologies. The evaluation was semi-automatic: consensus alignments were generated based on those produced by the participating systems, and the unique mappings found by each system were evaluated manually.
Multifarm (x8): This test case is based on a subset of the Conference data set, translated into ten different languages (Arabic, Chinese, Czech, Dutch, French, German, Italian, Portuguese, Russian, and Spanish) and the corresponding alignments between these ontologies. Results are evaluated against these alignments.
Interactive (x9): This test case offers the possibility to compare different matching tools which can benefit from user interaction. Its goal is to show if user interaction can improve matching results, which methods are most promising and how many interactions are necessary. Participating systems are evaluated on the conference data set using an oracle based on the reference alignment, which can generate erroneous responses to simulate user errors.
Instance matching (x10). The track aims at evaluating the performance of matching tools when the goal is to detect the degree of similarity between pairs of test formalism relations confidence modalities language benchmark
anatomy conference
largebio phenotype multifarm interactive
instance process model
OWL OWL OWL OWL OWL OWL OWL OWL OWL = = =, <= = = = =, <= = <=
items/instances expressed in the form of OWL Aboxes. Three independent tasks are defined: SABINE: The task is articulated in two sub-tasks called inter-lingual mapping and data linking. Both sub-tasks are based on OWL ontologies containing topics as instances of the class “Topic”. In inter-lingual mapping, two ontologies are given, one containing topics in the English language and one containing topics in the Italian language. The goal is to discover mappings between English and Italian topics. In data linking, the goal is to discover the DBpedia entity which better corresponds to each topic belonging to a source ontology.
SYNTHETIC: The task is articulated in two sub-tasks called UOBM and SPIMBENCH. In UOBM, the goal is to recognize when two OWL instances belonging to different data sets, i.e., ontologies, describe the same individual. In SPIMBENCH, the goal is to determine when two OWL instances describe the same Creative Work. Data Sets are produced by altering a set of original data. DOREMUS: The DOREMUS task contains real world data coming from the French National Library (BnF) and the Philharmonie de Paris (PP). Data are about classical music work and follow the DOREMUS model (one single vocabulary for both datasets). Three sub-tasks are defined called nine heterogeneities, four heterogeneities, and false-positive trap characterized by different degrees of heterogeneity in work descriptions.
Process Model Matching (x11): The track is concerned with the application of
ontology matching techniques to the problem of matching process models. It is based
on a data set used in the Process Model Matching Campaign 2015 [
Ontologies to be matched and (where applicable) reference alignments have been provided in advance during the period between June 1st and June 30th, 2016. This gave potential participants the occasion to send observations, bug corrections, remarks and other test cases to the organisers. The goal of this preparatory period is to ensure that the delivered tests make sense to the participants. The final test base was released on July 15th, 2016. The (open) data sets did not evolve after that. 2.4
During the execution phase, participants used their systems to automatically match the
test case ontologies. In most cases, ontologies are described in OWL-DL and serialised
in the RDF/XML format [
Participants were required to submit their wrapped tools by August 31st, 2016. Tools were then tested by the organisers and minor problems were reported to some tool developers, who were given the opportunity to fix their tools and resubmit them.
Initial results were provided directly to the participants between September 23rd and October 15th, 2016. The final results for most tracks were published on the respective pages of the OAEI website by October 15th, although some tracks were delayed.
The standard evaluation measures are usually precision and recall computed against the reference alignments. More details on the evaluation are given in the sections for the test cases. 2.6
Following the recent trend, the number of participating systems has remained approximately constant at slightly over 20 (see Figure 1). This year was no exception, as we counted 21 participating systems (out of 30 registered systems). Remarkably, participating systems have changed considerably between editions, and new systems keep emerging. For example, this year 10 systems had not participated in any of the previous OAEI campaigns. The list of participants is summarised in Table 2. Note that some systems were also evaluated with different versions and configurations as requested by developers (see test case sections for details). The goal of the benchmark data set is to provide a stable and detailed picture of each algorithm. For that purpose, algorithms are run on systematically generated test cases. The systematic benchmark test set is built around a seed ontology and many variations of it. Variations are artificially generated by discarding and modifying features from a seed ontology. Considered features are names of entities, comments, the specialisation hierarchy, instances, properties and classes. This test focuses on the characterisation of the behaviour of the tools rather than having them compete on real-life problems. Full description of the systematic benchmark test set can be found on the OAEI web site.
Since OAEI 2011.5, the test sets are generated automatically from different seed ontologies [20]. This year, we used two ontologies: biblio The bibliography ontology used in the previous years which concerns bibliographic references and is inspired freely from BibTeX; film A movie ontology developed in the MELODI team at IRIT (FilmographieV14). It uses fragments in French and labels in French and English.
The characteristics of these ontologies are described in Table 3.
The film data set was not available to participants when they submitted their systems. The tests were also blind for the organisers since we did not look into them before running the systems.
The reference alignments are still restricted to named classes and properties and use the “=” relation with confidence of 1. 4 https://www.irit.fr/recherches/MELODI/ontologies/
FilmographieV1.owl In order to avoid the discrepancy of last year, all systems were run in the most simple homogeneous setting. So, this year, we can write anew: All tests have been run entirely in the same conditions with the same strict protocol.
Evaluations were run on a Debian Linux virtual machine configured with four processors and 8GB of RAM running under a Dell PowerEdge T610 with 2*Intel Xeon Quad Core 2.26GHz E5607 processors and 32GB of RAM, under Linux ProxMox 2 (Debian). All matchers where run under the SEALS client using Java 1.8 and a maximum heap size of 8GB.
As a result, many systems were not able to properly match the benchmark. Evaluators availability is not unbounded and it was not possible to pay attention to each system as much as necessary.
Participation From the 21 systems participating to OAEI this year, only 10 systems were providing results for this track. Several of these systems encountered problems:
However we encountered problems with one very slow matcher (LogMapBio) that has been run anyway. RiMOM did not terminate, but was able to provide (empty) alignments for biblio, not for film. No timeout was explicitly set.
Reported figures are the average of 5 runs. As has already been shown in [20], there is not much variance in compliance measures across runs.
Compliance Table 4 synthesises the results obtained by matchers.
biblio
F-m.
Matcher
Prec.
Rec.
Prec.
film F-m.
Rec.
Systems that participated previously (AML, CroMatcher, Lily, LogMap, LogMapLite, XMap) still obtain the best results with Lily and CroMatcher still achieving an impressive .89 F-measure (against .90 and .88 last year). They combine very high precision (.96 and .97) with high recall (.83). The PhenoXX suite of systems return huge but poor alignments. It is surprising that some of the systems (AML, LogMapLite) do not clearly outperform edna (our edit distance baseline).
On the film data set (which was not known from the participants when submitting their systems, and actually have been generated afterwards), the results of biblio are fully confirmed: (1) those system able to return results were still able to do it besides CroMatcher and those unable, were still not able; (2) the order between these systems and their performances are commensurate. Point (1) shows that these are robust systems. Point (2) shows that the performances of these system are consistent across data sets, hence we are indeed measuring something. However, (2) has for exception LogMap and LogMapBio whose precision is roughly preserved but whose recall dramatically drops. A tentative explanation is that film contains many labels in French and these two systems rely too much on WordNet. Anyway, these and CroMatcher seem to show some overfit to biblio.
Polarity Besides LogMapLite, all systems have higher precision than recall as usual and usually very high precision as shown on the triangle graph for biblio (Figure 2). This can be compared with last year.
R=.9 R=.8 F=0.9
R=1. Pre=1fa.lign (b/f)
P=.9
Lily (b) P=.8
CroMatcher (b) R=.7
F=0.8
F=0.7 R=.6
F=0.6 R=.5
F=0.5
R=.4
R=.3 R=.2 R=.1 recall
Lily (f) XMap (f)
P=.7
P=.6
P=.5
P=.4
XMap (b)
LogMap (b) P=.3 LogMapLite (f)
AML (b) AML (f) P=.2
P=.1 precision
The precision/recall graph (Figure 3) confirms that, as usual, there are a level of recall unreachable by any system and this is where some of them go to catch their good F-measure.
Concerning confidence-weighted measures, there are two types of systems: those (CroMatcher, Lily) which obviously threshold their results but keep low confidence values and those (LogMap, XMap, LogMapBio) which provide relatively faithful measures. The former shows a strong degradation of the measured values while the latters resist
recall
refalign 1.00 AML 0.24 LogMap
0.38 PhenoMF 0.01 edna 0.50 XMap 0.40 LogMapLite
0.50 PhenoMM 0.01
Lily 0.82 CroMatcher
0.81 LogMapBio
0.19 PhenoMP 0.01 very well with XMap even improving its score. This measure which is supposed to reward systems able to provide accurate confidence values is beneficial to these faithful systems.
Speed Beside LogMapBio which uses alignment repositories on the web to find matches, all matchers do the task in less than 40 min (for biblio and 12h for film). There is still a large discrepancy between matchers concerning the time spent from less than two minutes for LogMapLite, AML and XMap to nearly two hours for LogMapBio (on biblio).
time film
stdev F-m./s.
Table 5 provides the average time, time standard deviation and 1/100e F-measure point provided per second by matchers. The F-measure point provided per second shows that efficient matchers are, like two years ago, LogMapLite and XMap followed by AML and LogMap. The correlation between time and F-measure only holds for these systems.
Time taken by systems is, for most of them, far larger on film than biblio and the deviation from average increased as well. This year, there is no increase or decrease of the performance of the best matchers which are roughly the same as previous years. Precision is still preferred to recall by the best systems. It seems difficult to other matchers to catch up both in terms of robustness and performances. This confirms the trend observed last year. 4
The anatomy test case confronts matchers with a specific type of ontologies from the biomedical domain. We focus on two fragments of biomedical ontologies which describe the human anatomy5 and the anatomy of the mouse6. This data set has been used since 2007 with some improvements over the years. 4.1
We conducted experiments by executing each system in its standard setting and we compare precision, recall, F-measure and recall+. The measure recall+ indicates the amount of detected non-trivial correspondences. The matched entities in a non-trivial correspondence do not have the same normalised label. The approach that generates only trivial correspondences is depicted as baseline StringEquiv in the following section.
We ran the systems on a server with 3.46 GHz (6 cores) and 8GB RAM allocated to each matching system. Further, we used the SEALS client to execute our evaluation. However, we slightly changed the way precision and recall are computed, i.e., the results generated by the SEALS client vary in some cases by 0.5% compared to the results presented below. In particular, we removed trivial correspondences in the oboInOwl namespace like:
http://...oboInOwl#Synonym = http://...oboInOwl#Synonym as well as correspondences expressing relations different from equivalence. Using the Pellet reasoner we also checked whether the generated alignment is coherent, i.e., that there are no unsatisfiable classes when the ontologies are merged with the alignment. 4.2
5 http://www.cancer.gov/cancertopics/cancerlibrary/
terminologyresources/ 6 http://www.informatics.jax.org/searches/AMA_form.shtml AML and XMap joined the track in 2013. DKP-AOM, Lily and CroMatcher participate for the second year in a row in this track. Lily participated in the track back in 2011. CroMatcher participated in 2013 but did not produce an alignment within the given time frame. Thus, this year we have 10 different systems (not counting different versions) which generated an alignment. For more details, we refer the reader to the papers presenting the systems.
Matcher AML CroMatcher XMap LogMapBio FCA Map LogMap LYAM Lily LogMapLite StringEquiv LPHOM Alin DKP-AOM-Lite DKP-AOM
Runtime
Size Precision
F-measure
Recall Recall+
Unlike the last two editions of the track when 6 systems generated an alignment in less than 100 seconds, this year only 4 of them were able to complete the alignment task in this time frame. These are AML, XMap, LogMap and LogMapLite. Similarly to the last 4 years LogMapLite has the shortest runtime, followed by LogMap, XMap and AML. Depending on the specific version of the systems, they require between 20 and 50 seconds to match the ontologies. The table shows that there is no correlation between quality of the generated alignment in terms of precision and recall and required runtime. This result has also been observed in previous OAEI campaigns.
The table also shows the results for precision, recall and F-measure. In terms of F-measure, the top 5 ranked systems are AML, CroMatcher, XMap, LogMapBio and FCA Map. LogMap is sixth with a F-measure very close to FCA Map. All the long-term participants in the track showed comparable results (in term or F-measure) to their last year’s results and at least as good as the results of the best systems in OAEI 2007-2010. LogMap and XMap generated the same number of correspondences in their alignment (XMap generated one correspondence more). AML and LogMapBio generated a slightly different number—16 correspondences more for AML and 18 less for LogMapBio.
The results for the DKP-AOM systems are identical this year; by contrast, last year the lite version performed significantly better in terms of the observed measures. While Lily had improved its 2015 results in comparison to 2011 (precision: from 0.814 to 0.870, recall: from 0.734 to 0.793, and F-measure: from 0.772 to 0.830), this year it performed similarly to last year. CroMatcher improved its results in comparison to last year. Out of all systems participating in the anatomy track CroMatcher showed the largest improvement in the observed measures in comparison to its values from the previous edition of the track.
Comparing the F-measures of the new systems, FCA Map (0.882) scored very close to one of the tracks’ long-term participants LogMap. Another of the new systems— LYAM—also achieved a good F-measure (0.869) which ranked sixth. As for the other two systems, LPHOM achieved a slightly lower F-measure than the baseline (StringEquiv) whereas Alin was considerably below the baseline.
This year, 9 out of 13 systems achieved an F-measure higher than the baseline which is based on (normalised) string equivalence (StringEquiv in the table). This is a slightly better result (percentage-wise) than last year’s (9 out of 15) and similar to 2014’s (7 out of 10). Two of the new participants in the track and the two DKP-AOM systems achieved an F-measure lower than the baseline. LPHOM scored under the StringEquiv baseline but at the same time it is the system that produced the highest number of correspondences. Its precision is significantly lower than the other three systems which scored under the baseline and generated only trivial correspondences.
This year seven systems produced coherent alignments which is comparable to the last two years, when 7 out of 15 and 5 out of 10 systems achieved this. From the five best systems only FCA Map produced an incoherent alignment. 4.3 Like for OAEI in general, the number of participating systems in the anatomy track this year was lower than in 2015 and 2013 but higher than in 2014, and there was a combination of newly-joined systems and long-term participants.
The systems that participated in the previous edition scored similarly to their previous results, indicating that no substantial developments were made with regard to this track. Of the newly-joined systems, (FCA Map and LYAM) ranked 4th and 6th with respect to the F-measure. 5
5.1
The conference test case requires matching several moderately expressive ontologies from the conference organisation domain.
The data set consists of 16 ontologies in the domain of organising conferences. These ontologies have been developed within the OntoFarm project7.
The main features of this test case are: – Generally understandable domain. Most ontology engineers are familiar with organising conferences. Therefore, they can create their own ontologies as well as evaluate the alignments among their concepts with enough erudition. – Independence of ontologies. Ontologies were developed independently and based on different resources, they thus capture the issues in organising conferences from different points of view and with different terminologies.
– Relative richness in axioms. Most ontologies were equipped with OWL DL axioms of various kinds; this opens a way to use semantic matchers.
Ontologies differ in their numbers of classes and properties, in expressivity, but also in underlying resources. 5.2
We provide results in terms of F-measure, comparison with baseline matchers and results from previous OAEI editions and precision/recall triangular graph based on sharp reference alignments. This year we can provide comparison between OAEI editions of results based on the uncertain version of reference alignment and on violations of consistency and conservativity principles.
ipants against blind reference alignments (labelled as rar2). This includes all pairwise combinations between 7 different ontologies, i.e., 21 alignments.
These reference alignments have been made in two steps. First, we have generated them as a transitive closure computed on the original reference alignments. In order to obtain a coherent result, conflicting correspondences, i.e., those causing unsatisfiability, have been manually inspected and removed by evaluators. The resulting reference alignments are labelled as ra2. Second, we detected violations of conservativity using the approach from [39] and resolved them by an evaluator. The resulting reference alignments are labelled as rar2. As a result, the degree of correctness and completeness of the new reference alignments is probably slightly better than for the old one. However, the differences are relatively limited. Whereas the new reference alignments are not open, the old reference alignments (labeled as ra1 on the conference web page) are available. These represent close approximations of the new ones.
Table 7 shows the results of all participants with regard to the reference alignment rar2. F0:5-measure, F1-measure and F2-measure are computed for the threshold that provides the highest average F1-measure. F1 is the harmonic mean of precision and recall where both are equally weighted; F2 weights recall higher than precision and F0:5 weights precision higher than recall. The matchers shown in the table are ordered according to their highest average F1-measure. We employed two baseline matchers. edna (string edit distance matcher) is used within the benchmark test case and with regard to performance it is very similar as the previously used baseline2 in the conference track; StringEquiv is used within the anatomy test case. This year these baselines divide matchers into two performance groups. The first group consists of matchers (CroMatcher, AML, LogMap, XMap, LogMapBio, FCA Map, DKP-AOM, NAISC and LogMapLite) having better (or the same) results than both baselines in terms of highest average F1-measure. Other matchers (Lily, LPHOM, Alin and LYAM) performed worse than both baselines. The performance of all matchers (except LYAM) regarding their precision, recall and F1-measure is visualised in Figure 4. Matchers are represented as squares or triangles. Baselines are represented as circles.
Further, we evaluated the performance of matchers separately on classes and properties. We compared the position of tools within overall performance groups and within only classes and only properties performance groups. We observed that while the position of matchers changed slightly in overall performance groups in comparison with CroMatcher
AML LogMap
XMap LogMapBio FCA Map DKP-AOM
NAISC edna LogMapLite StringEquiv
Lily LPHOM
Alin LYAM
Prec. F0:5-m. F1-m. F2-m. Rec. Inc.Align. Conser.V. Consist.V. only classes performance groups, a couple of matchers (DKP-AOM and FCA Map) worsen their position from overall performance groups with regard to their position in only properties performance groups due to the fact that they do not match properties at all (Alin and Lily also fall into this category). More details about these evaluation modalities are on the conference web page.
Comparison with previous years with regard to rar2 Seven matchers also participated in this test case in OAEI 2015. The largest improvement was achieved by CroMatcher (precision increased from .57 to .74 and recall increased from .47 to .65).
ues of all matches in the sharp reference alignments for the conference track are all 1.0. For the uncertain version of this track, the confidence value of a match has been set equal to the percentage of a group of people who agreed with the match in question (this uncertain version is based on the reference alignment labeled ra1). One key thing to note is that the group was only asked to validate matches that were already present in the existing reference alignments – so some matches had their confidence value reduced from 1.0 to a number near 0, but no new match was added.
There are two ways that we can evaluate matchers according to these “uncertain” reference alignments, which we refer to as discrete and continuous. The discrete evaluation considers any match in the reference alignment with a confidence value of 0.5 or greater to be fully correct and those with a confidence less than 0.5 to be fully incorrect. Similarly, a matcher’s match is considered a “yes” if the confidence value is greater than or equal to the matcher’s threshold and a “no” otherwise. In essence, this is the same as the “sharp” evaluation approach, except that some matches have been removed because less than half of the crowdsourcing group agreed with them. The continuous evaluation
F1-measure=0.7
F1-measure=0.6 F1-measure=0.5 AML CroMatcher DKP-AOM FCA-Map Lily LogMap LogMapBio LogMapLite LPHOM strategy penalises a matcher more if it misses a match on which most people agree than if it misses a more controversial match. For instance, if A B with a confidence of 0.85 in the reference alignment and a matcher gives that correspondence a confidence of 0.40, then that is counted as 0:85 0:40 = 0:34 of a true positive and 0:85{0:40 = 0:45 of a false negative.
Out of the 13 matchers, three (DKP-AOM, FCA-Map and LogMapLite) use 1.0 as the
confidence values for all matches they identify. Two of the remaining ten (Alin and
CroMatcher) have some variation in confidence values, though the majority are 1.0. The rest
of the systems have a fairly wide variation of confidence values. Last year, the majority
of these values were near the upper end of the [
Discussion When comparing the performance of the matchers on the uncertain reference alignments versus that on the sharp version, we see that in the discrete case all matchers performed slightly better. Improvement in F-measure ranged from 1 to 8 percentage points over the sharp reference alignment. This was driven by increased recall, which is a result of the presence of fewer “controversial” matches in the uncertain version of the reference alignment.
The performance of most matchers is similar regardless of whether a discrete or continuous evaluation methodology is used (provided that the threshold is optimised to achieve the highest possible F-measure in the discrete case). The primary exceptions
Alin
AML CroMatcher DKP-AOM FCA-Map
Lily
LogMap LogMapBio LogMapLite
LPHOM Light YAM++
NAISC XMap
Sharp Prec. F1-m. Rec.
Discrete Prec. F1-m. Rec.
Continuous
Prec. F1-m. Rec. to this are Lily and NAISC. These matchers perform significantly worse when evaluated using the continuous version of the metrics. In Lily’s case, this is because it assigns very low confidence values to some matches in which the labels are equivalent strings, which many crowdsourcers agreed with unless there was a compelling technical reason not to. This hurts recall, but using a low threshold value in the discrete version of the evaluation metrics “hides” this problem. NAISC has the opposite issue: it assigns relatively high confidence values to some matches that most people disagree with, such as “Assistant” and “Listener” (confidence value of 0.89). This hurts precision in the continuous case, but is taken care of by using a high threshold value (1.0) in the discrete case.
Seven matchers from this year also participated last year, and thus we are able to make some comparisons over time. The F-measures of all matchers either held constant or improved when evaluated against the uncertain reference alignments. Most matchers made modest gains (in the neighborhood of 1 to 6 percentage points). CroMatcher made the largest improvement, and it is now the second-best matcher when evaluated in this way. AgreementMakerLight remains the top performer.
Perhaps more importantly, the difference in the performance of most matchers between the discrete and continuous evaluation has shrunk between this year and last year. This is an indication that more matchers are providing confidence values that reflect the disagreement of humans on various matches.
performed evaluation based on detection of conservativity and consistency violations [39,40]. The consistency principle states that correspondences should not lead to unsatisfiable classes in the merged ontology; the conservativity principle states that correspondences should not introduce new semantic relationships between concepts from one of the input ontologies.
Table 7 summarises statistics per matcher. The table shows the number of unsatisfiable TBoxes after the ontologies are merged (Inc. Align.), the total number of all conservativity principle violations within all alignments (Conser.V.) and the total number of all consistency principle violations (Consist.V.).
Seven tools (Alin, AML, DKP-AOM, LogMap, LogMapBio, LPHOM and XMap) have no consistency principle violations (in comparison to five last year) and one tool (LYAM) generated only one incoherent alignment. There are two tools (Alin, LPHOM) that have no conservativity principle violations, and four more that have an average of only one conservativity principle violation (XMap, LogMap, LogMapBio and DKP-AOM). We should note that these conservativity principle violations can be “false positives” since the entailment in the aligned ontology can be correct although it was not derivable in the single input ontologies.
In conclusion, this year eight matchers performed better than both baselines on reference alignments which is not only consistent but also conservative. Further, this year seven matchers generated coherent alignments (against five matchers last year and four matchers the year before). This confirms the trend that increasingly matchers generate coherent alignments. Based on the uncertain reference alignments, more matchers are providing confidence values that reflect the disagreement of humans on various matches. 6
The largebio test case requires to match the large and semantically rich biomedical ontologies FMA, SNOMED-CT, and NCI, which contain 78,989, 306,591 and 66,724 classes, respectively. The test case has been split into three matching problems: FMA-NCI, FMA-SNOMED and SNOMED-NCI. Each matching problem has been further divided in 2 tasks involving differently sized fragments of the input ontologies: small overlapping fragments versus whole ontologies (FMA and NCI) or large fragments (SNOMED-CT).
The UMLS Metathesaurus [
Although the standard UMLS distribution does not directly provide alignments (in the sense of [21]) between the integrated ontologies, it is relatively straightforward to extract them from the information provided in the distribution files (see [25] for details).
It has been noticed, however, that although the creation of UMLS alignments combines expert assessment and auditing protocols they lead to a significant number of logical inconsistencies when integrated with the corresponding source ontologies [25].
Since alignment coherence is an aspect of ontology matching that we aim to promote, in previous editions we provided coherent reference alignments by refining the UMLS mappings using the Alcomo (alignment) debugging system [31], LogMap’s (alignment) repair facility [24], or both [26].
However, concerns were raised about the validity and fairness of applying automated alignment repair techniques to make reference alignments coherent [35]. It is clear that using the original (incoherent) UMLS alignments would be penalising to ontology matching systems that perform alignment repair. However, using automatically repaired alignments would penalise systems that do not perform alignment repair and also systems that employ a repair strategy that differs from that used on the reference alignments [35].
Thus, as of the 2014 edition, we arrived at a compromising solution that should be fair to all ontology matching systems. Instead of repairing the reference alignments as normal, by removing correspondences, we flagged the incoherence-causing correspondences in the alignments by setting the relation to “?” (unknown). These “?” correspondences will neither be considered as positive nor as negative when evaluating the participating ontology matching systems, but will simply be ignored. This way, systems that do not perform alignment repair are not penalised for finding correspondences that (despite causing incoherences) may or may not be correct, and systems that do perform alignment repair are not penalised for removing such correspondences.
To ensure that this solution was as fair as possible to all alignment repair strategies, we flagged as unknown all correspondences suppressed by any of Alcomo, LogMap or AML [?], as well as all correspondences suppressed from the reference alignments of last year’s edition (using Alcomo and LogMap combined). Note that, we have used the (incomplete) repair modules of the above mentioned systems.
The flagged UMLS-based reference alignment for the OAEI 2016 campaign is summarised in Table 9.
Reference alignment “=” corresp. “?” corresp.
FMA-NCI FMA-SNOMED SNOMED-NCI
We have run the evaluation on a Ubuntu Laptop with an Intel Core i7-4600U CPU @ 2.10GHz x 4 and allocating 15Gb of RAM. Precision, recall and F-measure have been computed with respect to the UMLS-based reference alignment. Systems have been ordered in terms of F-measure.
This year, out of the 21 systems participating in OAEI 2016, 13 were registered to participate in the largebio track, and 11 of these were able to cope with at least one of the largebio tasks within a 2 hour time frame. However, only 6 systems were able to complete more than one task, and only 4 systems completed all 6 tasks in this time frame. 6.3
Regarding the use of background knowledge, LogMap-Bio uses BioPortal as a mediating ontology provider, that is, it retrieves from BioPortal the most suitable top-10 ontologies for the matching task.
LogMap uses normalisations and spelling variants from the general (biomedical) purpose UMLS Lexicon.
AML has three sources of background knowledge which can be used as mediators between the input ontologies: the Uber Anatomy Ontology (Uberon), the Human Disease Ontology (DOID) and the Medical Subject Headings (MeSH).
System LogMapLite AML LogMap LogMapBio XMap FCA-Map Lily LYAM DKP-AOM DKP-AOM-Lite Alin # Systems
FMA-NCI Task 1 Task 2
FMA-SNOMED Task 3 Task 4
SNOMED-NCI
Task 5 Task 6
XMap uses synonyms provided by the UMLS Metathesaurus. Note that matching systems using UMLS Metathesaurus as background knowledge will have a notable advantage since the largebio reference alignment is also based on the UMLS Metathesaurus. Together with precision, recall, F-measure and run times we have also evaluated the coherence of alignments. We report (1) the number of unsatisfiabilities when reasoning with the input ontologies together with the computed alignments, and (2) the ratio of unsatisfiable classes with respect to the size of the union of the input ontologies.
We have used the OWL 2 reasoner HermiT [33] to compute the number of unsatisfiable classes. For the cases in which HermiT could not cope with the input ontologies and the alignments (in less than 2 hours) we have provided a lower bound on the number of unsatisfiable classes (indicated by ) using the OWL 2 EL reasoner ELK [27].
In this OAEI edition, only three distinct systems have shown alignment repair facilities: AML, LogMap and its LogMap-Bio variant, and XMap (which reuses the repair techniques from Alcomo [31]). Tables 11-12 (see last two columns) show that even the most precise alignment sets may lead to a huge number of unsatisfiable classes. This proves the importance of using techniques to assess the coherence of the generated alignments if they are to be used in tasks involving reasoning. We encourage ontology matching system developers to develop their own repair techniques or to use state-ofthe-art techniques such as Alcomo [31], the repair module of LogMap (LogMap-Repair) [24] or the repair module of AML [?], which have worked well in practice [26,22].
Table 10 shows which systems were able to complete each of the matching tasks in less than 24 hours and the required computation times. Systems have been ordered with respect to the number of completed tasks and the average time required to complete them. Times are reported in seconds.
The last column reports the number of tasks that a system could complete. For example, 8 system were able to complete all six tasks. The last row shows the number .
.
O
O
O 0 0 0
0 0 0 R
R
R D
D
D O
O
O 0 0 0 0
2 0 0 0 P
0 0 0 P 0 0 0
0 0 0
L X o A T M g
M M o a o p a L l
p 2 ,2 42 54 iTm 6 7 0
6 2 T 7 8 0 N
1 4 1 299 93 FP 2 225 28 FN .7 . O
L X o A T M g
M M o
o a p a L l
p 2 ,3 44 52 iTm 5 0 5 2
e 0 0 0 P .9 .9 .9 r 3 9 2 ec 3 4 9 . 0 0 0 R .7 . 0 0 0 P
Fig. 6. Average time between requests per task in the Conference data set (whiskers: Q1-1,5IQR, Q3+1,5IQR, IQR=Q3-Q1). The labels under the system names show the average number of requests and the mean time between the requests (calculated by taking the average of the average request intervals per task) for the ten runs and all tasks. tial set. LogMap and AML both request feedback on only selected mapping candidates (based on their similarity patterns or their involvement in unsatisfiabilities) and only present one mapping at a time to the user. XMap also presents one mapping at a time and asks mainly for true negatives. Only Alin employs the new feature in this year’s evaluation: analysing several conflicting mappings simultaneously, whereby a system can present up to three mappings together to the oracle, provided that each mapping presented has a mapped entity, i.e., class or property, in common with at least one other mapping presented.
The performance of the systems improves when interacting with a perfect oracle compared to no interaction. Although systems’ performance deteriorates when moving towards larger error rates there are still benefits from the user interaction—some of the systems’ measures stay above their non-interactive values even for the larger error rates. For the Anatomy track Alin detects only trivial correspondences in the non-interactive version while user interactions led to detecting some non-trivial correspondences.
The impact of the oracle’s errors is linear for Alin, AML and XMap and supra-linear for LogMap for all data sets. The ”Positive Precision” value affects the true positives and false positives, and the ”Negative Precision” value affects the true negatives and Fig. 7. Time between requests per task in the HP-MP data set (whiskers: Q1-1,5IQR, Q3+1,5IQR, IQR=Q3-Q1). The labels under the system names show the number of requests and the mean time between the requests. false negatives. The more a system relies on the oracle, the more sensitive it will be to its errors.
In general, XMap performs very few requests to the oracle compared to the other systems.
Two models for system response times are frequently used in the literature [
Regarding the number of unsatisfiable classes resulting from the alignments we observe some expected variations as the error increases. We note that, with interaction, the alignments produced by the systems are typically larger than without interaction, which makes the repair process harder. The introduction of oracle errors complicates the process further, and may make an alignment irreparable if the system follows the oracle’s feedback blindly. 10
The instance matching track aims at evaluating the performance of matching tools when the goal is to detect the degree of similarity between pairs of items/instances expressed in the form of RDF data. The track is organized in three independent tasks called SABINE, SYNTHETIC and DOREMUS. Each test is based on two data sets called source and target and the goal is to discover the matching pairs, i.e., mappings, among the instances in the source data set and the instances in the target data set. Fig. 9. Time between requests per task in the FMA-NCI data set (whiskers: Q1-1,5IQR, Q3+1,5IQR, IQR=Q3-Q1). The labels under the system names show the number of requests and the mean time between the requests.
For the sake of clarity, we split the presentation of task results in three different sections as follows.
SABINE is a modular benchmark in the domain of European politics for Social Business Intelligence (SBI) and it includes an ontology with 500 topics, both in English and Italian languages. The task is articulated in two sub-tasks called inter-lingual mapping and data linking.
In inter-lingual mapping, source and target datasets are OWL ontologies containing topics as instances of the class “Topic”. The source ontology contains topics in the English language; the target ontology contains other topics in the Italian language. The goal is to discover mappings between English and Italian topics by also defining the kind of relation which is most suitable for describing the discovered mapping between two matching topics.
In data linking, just the source dataset is defined and it is given to the participants as an OWL ontology containing topics as instances of the class “Topic”. The goal is to discover the best corresponding DBpedia entity for each topic in the source ontology.
The SABINE sub-tasks are defined as open tests, meaning that the set of expected
mappings, i.e., reference alignment, is given in advance to the participants and it
constitutes the gold standard for result evaluation. The task size is around 23K ontology
instances to consider. The gold standard has been defined through crowdsourcing
validation though the Argo system14. For creating the gold standard, workers are called to
recognize and confirm the mapping between instance topics of source and target
ontologies. In particular, a task is represented as a choice question in which a topic of the
source ontology is specified and a number of instance topics of the target ontology are
provided as possible mappings. A worker receiving a task to execute has to consider
a source topic and to choose the most appropriate mapping with a target topic among
those provided as possible options. Multi-worker task assignment and consensus
evaluation techniques are defined in Argo for quality assessment of the task result. A task is
assigned to a group G of 6 different workers. A group member autonomously executes
a task and independently produces the answer according to her/his personal feeling and
judgement. Given a task, its result is defined as an answer agreement, i.e., consensus,
among the members of the group that executed the task. Two workers agree on a task
result when they selected the same target topic as mapping with the given source topic.
14 http://island.ricerca.di.unimi.it/projects/argo/ (in Italian).
The mapping between the source and the target topics is confirmed and inserted in the
gold standard when the task answer having the highest degree of consensus within the
group G is supported by a qualified majority larger than 50%. Conversely, when a
qualified majority of workers is not found in G, the task is uncommitted and it is scheduled
for re-execution by a different group of workers with higher reliability. Further details
on the Argo techniques for task management are provided in [
Participants to the SABINE sub-tasks are LogMapIm, AML, LogMapLite, and RiMOM. Results are shown in Table 43. For each test, the tool performances are expressed in terms of precision, recall, and F-measure.
Inter-lingual mapping Data linkinging
Precision F-measure Recall Precision F-measure Recall LogMapIm AML LogMapLite RiMOM
We focus our considerations on AML and RiMOM that provided high-value results for precision, recall, and F-measure on both inter-lingual mapping and data linking subtasks. In particular, RiMOM outperforms AML on the inter-lingual mapping sub-task, in that both precision and recall values of RiMOM are higher than the corresponding values of AML. However, both tools are over 90% for precision and recall values, meaning that mapping corresponding instances of different languages is a successfully-addressed task by RiMOM and AML. In the data linking sub-task, AML outperforms RiMOM on precision and the difference between the tools on this result value is significant (i.e., AML >90% and RiMOM <50% on the precision value). On the opposite, for recall, we note that the RiMOM value is higher than the AML value, and the result of both tools is very positive (i.e., >85%). We argue that these results on the data linking sub-task are due to the problem of selecting the most appropriate mapping when a number of possible alternatives are available. Both AML and RiMOM are successful in providing a set of candidate DBpedia entities as target mapping with a given OWL instance (i.e., high recall value). On the opposite, the capability to choose/select the most appropriate mapping among the set of available options is still challenging and only AML succeeds in providing high-quality results on this task (i.e., high precision value).
UOBM and SPIMBENCH tasks are two of the evaluation tasks of instance matching tools where the goal is to determine when two OWL instances describe the same real world object. For the first task, the data sets have been produced by altering a set of source data and generated by SPIMBENCH [37] with the aim to generate descriptions of the same entity where value-based, structure-based and semantics-aware transformations are employed in order to create the target data. While, for the latter task the data sets have been generated with the University Ontology Benchmark (UOBM) [30] and transformed with the LANCE benchmark generator [36].
For both tasks, the transformations applied were a combination of value-based, structure-based, and semantics-aware test cases. The value-based transformations consider mainly typographical errors and different data formats, the structure-based transformations consider transformations applied on the structure of object and datatype properties and the semantics-aware transformations are transformations at the instance level considering the TBox information. The latter are used to examine if the matching systems take into account RDFS and OWL semantics in order to discover correspondences between instances that can be found only by considering information found in the TBox.
We stress that an instance in the source data set can have none or one matching counterpart in the target data set. A data set is composed of a TBox and a corresponding ABox. Source and target data sets share almost the same TBox (differences in the properties, due to the structure-based transformations). For SPIMBENCH, the sandbox scale is 10K triples 380 instances while the mainbox scale is 50K triples 1800 instances. We asked the participants to match the Creative Works instances (NewsItem, BlogPost and Programme) in the source data set against the instances of the corresponding class in the target data set. For UOBM, the sandbox scale is 14K triples 2.5K instances while the mainbox scale is 60K triples 10K instances. We asked the participants to match all the instances that are not common to the two data sets. For both tasks, we expected to receive a set of links denoting the pairs of matching instances that they found to refer to the same entity.
The participants to these tasks are LogMap, AML and RiMOM. For evaluation, we built a ground truth containing the set of expected links where an instance i1 in the source data set is associated with an instance in the target data set that has been generated as an altered description of i1.
The way that the transformations were done, was to apply value-based, structurebased and semantics-aware transformations, on different triples pertaining to one class instance.
The systems were judged on the basis of precision, recall and F-measure results that are shown in Tables 44 and 45.
Sandbox task Precision F-measure Recall
LogMap responds well regarding the SPIMBENCH task, while the performance drops when matching the data sets of the UOBM task. LogMap is automatic and does not require the definition of a configuration file in contrast to AML and RiMOM. LogMap AML RiMOM
AML responds well regarding the SPIMBENCH task, while the performance drops when matching the data sets of the UOBM task. AML had to turn off the reasoner in order to handle missing information about the domain and range of TBox properties.
LogMap and AML produce links that are quite often correct (resulting in a good precision) but fail in capturing a large number of the expected links (resulting in a lower recall).
RiMOM performs better than any other system for most of the tasks; it performs excellent in the case of SPIMBENCH but, although it exhibits the best results for the Sandbox track of UOBM, its performance drops for the Mainbox track. For RiMOM, the probability of capturing a correct link is high, but the probability of a retrieved link to be correct is lower, resulting in a high recall but not a high precision.
The main comments for the SPIMBENCH and UOBM tasks are: – LogMap and AML have consistent behaviour regarding Sandbox and Mainbox. – RiMOM has a consistent behaviour for the SPIMBENCH task and an inconsistent behaviour for the UOBM task. – All systems performed well for the SPIMBENCH task. – The UOBM data sets seem to be more “difficult” for both IM systems, and this difficulty stems from the data set itself, rather than from the transformations imposed by LANCE. – The UOBM data sets seem to be more difficult for both IM systems, and this difficulty stems from the data set itself, rather than from the transformations imposed by LANCE. In particular, an important source of difficulty for the systems is that the URIs of the instances in the data set look very similar to each other, so even the change of a URI can lead to false positives or false negatives.
The DOREMUS task, having its premier at OAEI, contains real world data sets coming
from two major French cultural institutions—The BnF (French National Library) and
the PP (Philharmonie de Paris). The data are about classical music works and follow the
DOREMUS model (one single vocabulary for both data sets) issued from the
DOREMUS project15. Each data entry, or instance, is a bibliographical record about a musical
piece, containing properties such as the composer, the title(s) of the work, the year of
creation, the key, the genre, the instruments, to name a few. These data have been
converted to RDF from their original UNI- and INTER-MARC format and anchored to the
DOREMUS ontology and a set of domain controlled vocabularies by the help of the
marc2rdf converter16, developed for this purpose within the DOREMUS Project (for
15 http://www.doremus.org
16 https://github.com/DOREMUS-ANR/marc2rdf
more details on the conversion method and on the ontology we refer to [
Nine heterogeneities. This task consists in aligning two small data sets, BnF-1 and PP1, containing about 40 instances each, by discovering 1:1 equivalence relations between their instances. There are 9 types of heterogeneities that these data manifest, that have been identified by the music library experts, such as multilingualism, differences in catalogues, differences in spelling, different degrees of description (number of properties). Four heterogeneities. This task consists in aligning two larger data sets, BnF-2 and PP-2, containing about 200 instances each, by discovering 1:1 equivalence relations between the instances that they contain. There are 4 types of heterogeneities that these data manifest, that we have selected from the nine in Task 1 and that appear to be the most problematic: 1) Orthographical differences, 2) Multilingual titles, 3) Missing properties, 4) Missing titles.
The False Positives Trap. This task consists in correctly disambiguating the instances contained in two data sets, BnF-3 and PP-3, by discovering 1:1 equivalence relations between the instances that they contain. We have selected several groups of pairs of works with highly similar descriptions where there exists only one correct match in each group. The goal is to challenge the linking tools capacity to avoid the generation of false positives and match correctly instances in the presence of highly similar but still distinct candidates.
AML (th=0.2) AML (th=0.6) RiMOM
9 heterogeneities
Prec. F-m. Rec.
Results Only two systems returned results on the track: AML and RiMOM. Note that
AML has been configured with two different thresholds. The results of their
performances, evaluated by using precision, recall and F-measure, on each of the three tasks
can be seen in Table 46. The best performance in terms of F-measure is provided by the
AML tool with a threshold of 0:2 on all tasks.
11
In 2013 and in 2015 the community interested in business process modelling conducted
an evaluation campaign similar to OAEI [
Only few systems have been marked as capable of generating alignments for the Process Model Matching track. We have tried to execute all these systems, however, some of them generated only trivial TBox mappings instead of mappings between activities. After contacting the developer of the systems, we received the feedback that the systems have been marked mistakenly and are designed for terminological matching only. We have excluded them from the evaluation. Moreover, we tried to run all systems that were marked as instance matching tools, which have been submitted as executable SEALS bundles. One of these tools (LogMap), generated meaningful results and was also added to the set of systems that we evaluated. Finally we evaluated three systems (AML, LogMap, and DKP), one of these systems was configured in two different settings related to the treatment of events-to-activity mappings. This was the tool DKP. Thus we distinguish between DKP and DKP*.
In our evaluation, we computed standard precision and recall, as well as the
harmonic mean known as F-measure. The data set we used consists of several test cases.
We aggregated the results and present the micro average results. The gold standard we
used for our first set of evaluation experiments is based on the gold standard that has
also been used at the Process Model Matching Contest in 2015 [
Moreover, we extended our evaluation (“Standard” in Table 47) by a new evaluation measure that makes use of a probabilistic reference alignment (“Probabilistic” in Table 47). This probabilistic measure is based on a gold standard which is manually and independently generated by several domain experts. The number of votes of these annotators are applied as support values in the probabilistic evaluation. For a detailed discussion, please refer to [28]. 11.2 Table 47 summarises the results of our evaluation. “P” abbreviates precision, “R” is recall, “FM” stands for F-measure and “Rk” means rank. The prefix “Pro” indicates the probabilistic versions of the precision, recall, F-measure and the associated rank. These metrics are explained below. Participants of the Process Model Matching Contest in 2015 (PMMC 2015) are depicted in grey font, while OAEI 2016 participants are shown in black font. The OAEI participants are ranked on position 1, 8, 9 and 11 with an overall number of 16 systems listed in the table (when using the standard metrics). Note that AML-PM at the PMMC 2015 was a matching system that was based on a predecessor of AML participating at OAEI 2016. The good results of AML are surprising, since we expected that matching systems specifically developed for the purpose of process model matching would outperform ontology matching systems applied to the special case of process model matching. While AML contains also components that are specifically designed for the process matching task (a flooding-like structural matching algorithm), its relevant main components are components developed for ontology matching and the sub-problem of instance matching.
R FM
AML OAEI-16 221 AML-PM PMMC-15 579 BPLangMatch PMMC-15 277 DKP OAEI-16 177 DKP* OAEI-16 150 KnoMa-Proc PMMC-15 326 KMatch-SSS PMMC-15 261 LogMap OAEI-16 267 Match-SSS PMMC-15 140 OPBOT PMMC-15 234 pPalm-DS PMMC-15 828 RMM-NHCM PMMC-15 220 RMM-NLM PMMC-15 164 RMM-SMSL PMMC-15 262 RMM-VM2 PMMC-15 505 TripleS PMMC-15 230 0,719 0,685 0,702 1 0,742 0,283 0,269 0,672 0,385 14 0,377 0,398 0,368 0,440 0,401 12 0,532 0,272 0,621 0,474 0,538 8 0,686 0,219 0,680 0,440 0,534 9 0,772 0,211 0,337 0,474 0,394 13 0,506 0,302 0,513 0,578 0,544 6 0,563 0,274 0,449 0,517 0,481 11 0,594 0,291 0,807 0,487 0,608 4 0,761 0,192 0,603 0,608 0,605 5 0,648 0,258 0,162 0,578 0,253 16 0,210 0,335 0,691 0,655 0,673 2 0,783 0,297 0,768 0,543 0,636 3 0,681 0,197 0,511 0,578 0,543 7 0,516 0,242 0,216 0,470 0,296 15 0,309 0,294 0,487 0,483 0,485 10 0,486 0,210 0,410 0,387 0,360 0,333 0,331 0,378 0,368 0,390 0,307 0,369 0,258 0,431 0,306 0,329 0,301 0,293
In the probabilistic evaluation, however, the OAEI participants gain position 2, 3, 9 and 10, respectively. LogMap rises from position 11 to 3. The (probabilistic) precision improves over-proportionally for this matcher, because LogMap generates many correspondences which are not included in the binary gold standard but are included in the probabilistic one. The ranking of LogMap demonstrates that a strength of the probabilistic metric lies in the broadened definition of the gold standard where weak mappings are included but softened (via the support values).
The lessons learned from running OAEI 2016 were the following: A) This year, as suggested in previous campaigns, we requested tool registration in June and preliminary submission of wrapped systems by the end of July. This measure was successful in reducing the number of systems with errors and incompatibilities with the SEALS client during the evaluation phase as had happened in the past. However, not all systems complied with the deadlines, and some did have problems, which still delayed the evaluation. In future editions, we must be more strict in enforcing the participation protocol.
B) Thanks in part to the new submission schedule, this marked the first OAEI edition where all participants and all tracks were evaluated using the SEALS client. Nevertheless, some system developers still struggled to get their systems working with the client, mostly due to incompatible versions of libraries. This recurring problem, plus the effort required to update the SEALS client’s libraries, lead to the consideration of whether it would not be better to develop a simpler, more streamlined evaluation solution. (c) Probabilistic F-measure
C) The continued absence of the SEALS web portal did not seem to affect participation, as the Google drive solution for submission was well received by the participants. OAEI may move towards a cloud-based solution.
D) While the number of participants this year was similar to that of recent years, their distribution through the tracks was uneven. Long-standing tracks had no shortage of participants, but alas the same was not true for the Interactive, Process Model (new) or Instance (new data sets) tracks. One reason for this is that the OAEI data sets have been released too close to the submission deadline to allow system developers to develop their systems to tackle them all—the timing is barely sufficient to allow serious development focusing on one new data set. Thus, with prize money on offer on one of the new tracks, it is no surprise that system developers were polarised towards that track and eschewed the other new ones. We should consider anticipating the deadline for initial release of OAEI data sets, particular for those that are new, in order to give system developers more time to tackle them, thereby increasing participation.
E) The increasing variety of OAEI tracks also poses difficulties to system developers in configuring their systems to handle different types of tasks. It is noteworthy that only two systems, both of which are long-term OAEI participants, have tackled all tracks—and one of them did so using external configuration files specifying the type of task. One solution to facilitate participation in multiple tracks would be to have the evaluation client transmit to the system the specifications of the task, e.g., whether classes, properties, and/or individuals are to be matched, and whether only a specific subset of them are to be matched. This would also make the tasks more realistic, in the sense that in normal use, a user would provide to the ontology matching system this type of information.
F) With regard to the low participation in the Process Model and Instance tracks, it merits considering whether enforcing adherence to the SEALS client and ontologybased data sets were not deterrent factors. It should be noted that the Process Model Matching Contest (PMMC) received a much larger number of participants in 2015 than did the Process Model track, and that there is a considerable number of publications on data interlinking systems, but only one of these participated in the Instance track.
G) In previous years we identified the need for considering non-binary forms of evaluation, namely in cases where there is uncertainty about some of the reference mappings. A first non-binary evaluation type was implemented in last year’s Conference track, but this year two new tracks followed suit: Disease and Phenotype where the evaluation was semantic, and Process Model, where it was probabilistic. These new strategies should provide a fairer evaluation of the systems in complex test cases.
The lessons learned in the various OAEI 2016 track were the following: largebio: While the current reference alignments, with incoherence-causing mappings flagged as uncertain, make the evaluation fair to all systems, they are only a compromise solution, not an ideal one. Thus, we should aim for manually repairing and validating the reference alignments for future editions. phenotype: The prize offered in this track, thanks to the kind sponsorship of the Pistoia Alliance Ontologies Mapping project, was positively accepted by the community and helped attract new participants. However, it also had a polarising effect, with some systems focusing exclusively in this track. In future editions, we will consider including a prize across OAEI tracks in order to motivate developers to successfully participate in more than one track. interactive: The new functionality of the Oracle allowing systems to submit a set of up to three conflicting mappings, rather than a mapping at a time, was successfully exploited by one new participating system. Nevertheless, this track’s participation has remained low, as most systems participating in OAEI focussed exclusively on fully automatic matching. We hope to draw more participants to this track in the future and will continue to expand it so as to better approximate real user interactions. process model: The results of the new Process Model track have shown that the participating ontology matching systems are capable of generating very good results for the specific problem of process model matching. This shows that the basic components of an ontology matching system can also be successfully applied to other kind of matching problems. instance: In order to attract more instance matching systems to participate in value semantics (val-sem), value structure (val-struct), and value structure semantics (valstruct-sem) tasks, we need to produce benchmarks that have fewer instances (in the order of 10000), of the same type (in our benchmark we asked systems to compare instances of different types). To balance those aspects, we must then produce data sets with more complex transformations. 13 OAEI 2016 saw the same number (21) of participants as in recent years, with a healthy mix of new and returning systems. While some new participants were mainly drawn by the allure of prize money in the new Disease and Phenotype track, the very fact that there was prize money on offer shows that interest in ontology matching is not waning, which bodes well for the future of OAEI. All the test cases were performed on the SEALS client, including those in the instance matching track, which is good news regarding the interoperability of matching systems. Furthermore, the fact that the SEALS client can be used for such a variety of tasks is a good sign of its relevance.
Unlike previous years, this year there was no noticeable improvement with regard to system run times—for instance, the distribution of run times in Anatomy and Large Biomedical Ontologies was approximately the same as last year. There was also no progress with regard to the ability to handle large ontologies and data sets, as the number of systems able to cope with the Large Biomedical Ontologies data set in full was the same as last year, and all systems able to cope with the Instance Synthetic data set were established systems already known for their ability to handle large data sets. Finally, there was no progress with regard to alignment repair systems, with only a few returning systems employing them. As a consequence, incoherent alignments are common.
With regard to F-measure, some returning systems showed substantial improvements, but overall, the improvements in F-measure were subtle in Anatomy and Large Biomedical Ontologies, and non-existent in Conference. As has been the trend, most systems favour precision over recall.
Most of the participants have provided a description of their systems and their experience in the evaluation. These OAEI papers, like the present one, have not been peer reviewed. However, they are full contributions to this evaluation exercise and reflect the hard work and clever insight people put into the development of participating systems. Reading the papers of the participants should help people involved in ontology matching find out what makes these algorithms work and what could be improved.
The Ontology Alignment Evaluation Initiative will strive to continue to be a reference to the ontology matching community by improving both the test cases and the testing methodology to better reflect the actual needs of the community. Evaluating ontology matching systems remains a challenging but critical topic, which is essential to enable the progress of this field [38]. More information can be found at:
We warmly thank the participants of this campaign. We know that they have worked hard to have their matching tools executable in time and they provided useful reports on their experience. The best way to learn about the results remains to read the papers that follow.
We would also like to thank the Pistoia Alliance9 which sponsored the Disease and Phenotype track and funded the prize for the winners.
We are very grateful to the Universidad Polite´cnica de Madrid (UPM), especially to Nandana Mihindukulasooriya and Asuncio´n Go´mez Pe´rez, for moving, setting up and providing the necessary infrastructure to run the SEALS repositories.
We are also grateful to Martin Ringwald and Terry Hayamizu for providing the reference alignment for the anatomy ontologies and thank Elena Beisswanger for her thorough support on improving the quality of the data set.
We thank Khiat Abderrahmane for his support in the Arabic data set and Catherine Comparot for her feedback and support in the MultiFarm test case.
We also thank for their support the other members of the Ontology Alignment Evaluation Initiative steering committee: Yannis Kalfoglou (Ricoh laboratories, UK), Miklos Nagy (The Open University (UK), Natasha Noy (Stanford University, USA), Yuzhong Qu (Southeast University, CN), York Sure (Leibniz Gemeinschaft, DE), Jie Tang (Tsinghua University, CN), George Vouros (University of the Aegean, GR).
Michelle Cheatham has been supported by the National Science Foundation award ICER1440202 “EarthCube Building Blocks: Collaborative Proposal: GeoLink”.
Je´roˆme Euzenat, Ernesto Jimenez-Ruiz, Christian Meilicke, Heiner Stuckenschmidt and Ca´ssia Trojahn dos Santos have been partially supported by the SEALS (IST-2009-238975) European project in previous years.
Daniel Faria was supported by the ELIXIR-EXCELERATE project (INFRADEV-3-2015).
Ernesto Jimenez-Ruiz has also been partially supported by the Seventh Framework Program (FP7) of the European Commission under Grant Agreement 318338, “Optique”, the EPSRC projects DBOnto and ED3, the Research Council of Norway project BigMed, and the Centre for Scalable Data Access (SIRIUS).
Catia Pesquita was supported by the FCT through the LASIGE Strategic Project (UID/CEC/00408/2013) and the research grant PTDC/EEI-ESS/4633/2014.
Ondrˇej Zamazal has been supported by the CSF grant no. 14-14076P. 26. Ernesto Jime´nez-Ruiz, Christian Meilicke, Bernardo Cuenca Grau, and Ian Horrocks. Evaluating mapping repair systems with large biomedical ontologies. In Proc. 26th Description Logics Workshop, 2013. 27. Yevgeny Kazakov, Markus Kro¨tzsch, and Frantisek Simancik. Concurrent classification of EL ontologies. In Proc. 10th International Semantic Web Conference (ISWC), Bonn (DE), pages 305–320, 2011. 28. Elena Kuss, Henrik Leopold, Han Van der Aa, Heiner Stuckenschmidt, and Hajo A. Reijers.
Probabilistic evaluation of process model matching techniques. In Lecture notes in computer science. Conceptual modeling: 35th international conference, ER 2016, Gifu, Japan, November 14-17, 2016, pages 279–292, 2016. 29. Pasquale Lisena, Manel Achichi, Eva Ferna´ndez, Konstantin Todorov, and Raphae¨l Troncy.
Exploring linked classical music catalogs with overture. In ISWC PD: International Semantic Web Conference Posters and Demos, 2016. 30. L. Ma, Y. Yang, Z. Qiu, G, Xie, Y. Pan, and S. Liu. Towards a Complete OWL Ontology
Benchmark. In ESWC, 2006. 31. Christian Meilicke. Alignment Incoherence in Ontology Matching. PhD thesis, University
Mannheim, 2011. 32. Christian Meilicke, Rau´l Garc´ıa Castro, Frederico Freitas, Willem Robert van Hage, Elena Montiel-Ponsoda, Ryan Ribeiro de Azevedo, Heiner Stuckenschmidt, Ondrej Sva´b-Zamazal, Vojtech Sva´tek, Andrei Tamilin, Ca´ssia Trojahn, and Shenghui Wang. MultiFarm: A benchmark for multilingual ontology matching. Journal of web semantics, 15(3):62–68, 2012. 33. Boris Motik, Rob Shearer, and Ian Horrocks. Hypertableau reasoning for description logics.
Journal of Artificial Intelligence Research, 36:165–228, 2009. 34. Heiko Paulheim, Sven Hertling, and Dominique Ritze. Towards evaluating interactive ontology matching tools. In Proc. 10th Extended Semantic Web Conference (ESWC), Montpellier (FR), pages 31–45, 2013. 35. Catia Pesquita, Daniel Faria, Emanuel Santos, and Francisco Couto. To repair or not to repair: reconciling correctness and coherence in ontology reference alignments. In Proc. 8th ISWC ontology matching workshop (OM), Sydney (AU), pages 13–24, 2013. 36. Tzanina Saveta, Evangelia Daskalaki, Giorgos Flouris, Irini Fundulaki, Melanie Herschel, and Axel-Cyrille Ngonga Ngomo. Lance: Piercing to the heart of instance matching tools.
In International Semantic Web Conference, pages 375–391. Springer, 2015. 37. Tzanina Saveta, Evangelia Daskalaki, Giorgos Flouris, Irini Fundulaki, Melanie Herschel, and Axel-Cyrille Ngonga Ngomo. Pushing the limits of instance matching systems: A semantics-aware benchmark for linked data. In WWW, Companion Volume, 2015. 38. Pavel Shvaiko and Je´roˆme Euzenat. Ontology matching: state of the art and future challenges.
IEEE Transactions on Knowledge and Data Engineering, 25(1):158–176, 2013. 39. Alessandro Solimando, Ernesto Jime´nez-Ruiz, and Giovanna Guerrini. Detecting and correcting conservativity principle violations in ontology-to-ontology mappings. In The Semantic Web–ISWC 2014, pages 1–16. Springer, 2014. 40. Alessandro Solimando, Ernesto Jimenez-Ruiz, and Giovanna Guerrini. Minimizing conservativity violations in ontology alignments: Algorithms and evaluation. Knowledge and Information Systems, 2016. 41. York Sure, Oscar Corcho, Je´roˆme Euzenat, and Todd Hughes, editors. Proc. ISWC Workshop on Evaluation of Ontology-based Tools (EON), Hiroshima (JP), 2004.