ALIN is an ontology alignment system specialized in the interactive alignment of ontologies. Its main characteristic is the selection of correspondences to be shown to the expert, depending on the previous feedbacks given by the expert. This selection is based on semantic and structural characteristics. ALIN has obtained the alignment with the highest quality in the interactive tracking for Conference data set. This paper describes its configuration for the OAEI 2016 competition and discusses its results.
1.1
ALIN is an ontology alignment system, specialized in the ontology interactive alignment, based primarily on linguistic matching techniques, using the Wordnet as external resource. After generating an initial set of correspondences ( called set of candidate correspondences, which are the correspondences selected to receive the feedback from the expert ), interactions are made with the expert, and to each interaction, the set of candidate correspondences is modified. The modification of the set of candidate correspondences is through the use of the structural analysis of ontologies and use of alignment anti-patterns. The interactions continue until there are no more candidate correspondences left. ALIN was built with a special focus on the interactive matching track of OAEI 2016. 1.2
The ALIN workflow is shown in figure 1. 1. Load of the ontologies with load of classes, object properties and data properties through the Align API1. For each entity some data are stored such as name and label. In the case of classes, their superclasses and disjunctions are saved. In the case of object properties are saved the properties that are their hypernyms and their associated classes. The classes of property data are saved, too. ALIN does not use instances. After loading, the matching problem is profiled taking into account the size of the ontologies. The ALIN can only work with ontologies whose entity names are in English. 2. As an initial set of candidate correspondences a stable marriage algorithm with incomplete preference lists with maximum size of the list equals to 1, using linguistic metrics to sort the priority list was used [2]. The list is sorted in decreasing order. For this algorithm only the correspondence whose first entity is in the list of second entity and vice-versa is selected. The linguist metrics used are Jaccard, JaroWinkler and n-Gram [3] provided by Simmetrics API2 and Wu-Palmer, Jiang-Conrath and Lin [3] provide by ws4j API3 that use Wordnet. To use Wordnet the canonical form of the word is needed, therefore Stanford CoreNLP API4 was considered. The algorithm is run six times, once by each metric, and the result set is the union of results of each metric. 3. The value of the similarity metrics ( Wu-Palmer, Jiang-Conrath, Lin, Jaccard, Jaro-Winkler and n-Gram ) vary from 0 to 1 ( 1 is the maximum value ). When a correspondence in the set of candidate correspondences has all the six metrics with the maximum value, it is added to the final alignment and removed from the set of candidate correspondences. There are exceptions to this rule, some correspondences that fall into some structural patterns are not put on the final alignment and are not removed from the set of candidate correspondences. 4. The correspondences whose entities are not in the same synset of wordnet are removed from the set of candidate correspondences. These correspondences are put into a backup set, and can return to the set of candidate correspondences using structural analysis. 5. At this point the interactions with the expert begin. The correspondences in the set of candidate correspondences are sorted by the sum of similarity metric values, with the greatest sum first. The options are showed one by one to the expert. The first correspondence is showed and it is removed from the list after the answer of the expert. The set of candidate correspondences has, at first, only correspondences of classes. When the expert answer one question, the set of candidate correspondences is changed. Correspondences ( besides the 1 2 3 “ A lignment API ” . Available at http://alignapi.gforge.inria.fr/ Last accessed on Apr, 11, 2016. “ S tring Similarity Metrics for Information Integration ” . Available on http://www.coli.uni-saarland.de/courses/LT1/2011/slides/stringmetrics.pdf. Last accessed on Apr, 19, 2016. “ W S4J ” . Available at https://code.google.com/archive/p/ws4j/ Last accessed on Apr, 11, 2016. “ S tanford CoreNLP ” . Available at http://stanfordnlp.github.io/CoreNLP/ Last accessd on Sept, 15, 2016. correspondence answered by expert ) can be removed and included, depending on the answer of the expert. If the expert does not accept the correspondence it is removed from the set of candidate correspondences. But if the expert accepts the correspondence it is removed from the set of candidate correspondences and put in the final alignment.
At each interaction with the specialist we also:
- We remove from the set of candidate correspondences and disregard all the
correspondences that are in anti-pattern of alignment [
This step continues until the set of candidate correspondences is empty. 1.3
ALIN is available through Mediafire (https://www.mediafire.com/folder/726zohj792kod/ALIN) as a package for running through the SEALS client. 2
The system ALIN has been developed with its focus on interactive ontology alignment. The approach performs better when the number of data and object properties is proportionately large. ALIN considers properties associated to correspondent classes when selecting entities for user feedback, thus allowing for increased recall. When the number of properties in the ontologies is small, the system still generates a very precise alignment, but its recall tends to decrease. Another characteristic of ALIN is its reliance on an interactive phase. The noninteractive phase of the system is quite simple, mainly based on maximum string similarity, specializing in maintaining a high precision without worrying about recall, generating initially a low f-measure. The recall increases in the interactive phase. Finally, ALIN is also not robust to users errors. The system uses a number of techniques that take advantage of the expert response to reach other conclusions when the expert gives a wrong answer it is propagated generating other errors, thereby diminishing the f-measure. 2.1
As expected the participation of ALIN in non-interactive alignment processes showed the following results: high precision and not so high recall, as can be seen in Table 1, where recall+ field refers to non-trivial correspondences found and Coherent field filled by + indicates that the generated alignment is consistent.
Regarding the Conference track, as ALIN evaluates only the properties associated with classes already evaluated as belonging to the alignment, the alignment of the M2 type (which take into account only the properties of ontologies) were with the fmeasure = 0, as can be seen in Table 3. As properties are evaluated only in the interactive phase in the ALIN, alignments of type M1 (only classes) remained with a higher recall than M3 (classes and properties), as can be seen in Tables 2 and 4, because the reference alignments of type M3 contain properties besides classes. 2.2
In this track the program ALIN showed the highest precision among the four evaluated tools when the error rate is zero. When the error rate increases both the precision as the recall falls, reducing the f-measure. This is expected and explained earlier.
Table 5. - Participation of ALIN in interactive alignment - Anatomy track.
Table 6. - Participation of ALIN in interactive alignment - Conference track. As ontologies of the Anatomy Track contains almost no properties, techniques used in ALIN can not be utilized, the selection of properties associated with classes assessed as belonging to the alignment, this has limited the increase in recall, which influenced the f-measure, as can be seen in Table 5.
In this track ALIN stood out, showing the greatest f-measure among the four tools when the error rate is zero, as with a loss of f-measure when the error rate increases, as can be seen in Table 6. 3
Evaluating the results it can be seen that the system can be improved towards: (a) handling user error rate; (b) generating a higher quality (especially w.r.t. recall) initial alignment in its non-interactive phase; (c) reducing the number of interactions with the expert; and (d) optimize the process to reduce its execution time. 4
Within certain characteristics, the ALIN system stands out in ontology alignment process in interactive application scenarios, especially when the amount of data and object properties are also subject to the alignment and when the expert does not make mistakes. With these features there is an alignment generated with relatively high precision and recall.