=Paper=
{{Paper
|id=Vol-1963/paper634
|storemode=property
|title=Efficient Synonym Search by Semantic Linking of Multiple Data Sets
|pdfUrl=https://ceur-ws.org/Vol-1963/paper634.pdf
|volume=Vol-1963
|authors=Kenny Knecht,Bérénice Wulbrecht,Filip Pattyn,Hans Constandt
|dblpUrl=https://dblp.org/rec/conf/semweb/KnechtWPC17
}}
==Efficient Synonym Search by Semantic Linking of Multiple Data Sets==
https://ceur-ws.org/Vol-1963/paper634.pdf
Efficient synonym search by semantic linking of
multiple data sets
Kenny Knecht1 Bérénice Wulbrecht1 Filip Pattyn1 Hans Constant1
ONTOFORCE NV, Gent, Belgium,
kenny@ontoforce.com
Abstract. We describe a method to automatically pick a highly relevant
subset of synonyms to broaden a text search based on keywords.
Public datasets in the bio-medical area tend to provide a plethora of
synonyms or alternative names. It is not uncommon that chemicals or
diseases have more than 50 different alternative names in data sets like
UMLS or ChEMBL. This may result in inefficient searches and some-
times even in false positives if you use these to extend an initial search.
Through semantic linking of several datasets we define a heuristic which
increases the power of the search meanwhile making it more efficient. We
evaluated the method on the 500 most common keyword searches used
the first 6 months of 2017 in the semantic web platform DISQOVER
(www.disqover.com). More than 98% of the hits are retrieved back by
submitting only 16% of the synonyms.
We implemented this method as a visual suggestion, which the user can
override manually at any time. Notwithstanding the fact that we focus
our examples and concrete implementation on the biomedical databases
in the publicly available DISQOVER, we would like to stress that the
method is much more generally applicable.
Keywords: semantic web, text search, synonyms, data integration
1 The problem
Traditionally search engines start with a text query. All the documents contain-
ing that text are subsequently returned: in DISQOVER this can be publications,
clinical trials or funded research programs, but also other concepts like diseases,
genes, variants or other chemicals.
The central example will be all documents about the concept aspirin. Merely
typing ”aspirin” will surely return a lot of relevant results, but some will also
be missed. For example some documents may only mention the more scientific
name ”Acetylsalicylic acid”. So logically the user would like to expand the search
with synonyms to broaden his or her search, i.e. query for ”aspirin” OR ”Acetyl-
salicylic acid”.
Since DISQOVER brings together many data sources -many of them actu-
ally contributing synonyms and other alternative names- this can be easily au-
tomated. The record for Aspirin for example collects data from no less than 13
2
different databases (HMDB, DrugCentral, DrugBank, HSDB, UNII, ChEMBL,
UMLS, ChEBI, IUPHAR Compendium, SureChEMBL, RxNorm, MeSH, Pub-
Chem ), from which 8 contribute to alternative names. In total the public
databases gives 75 distinct alternative names for aspirin, many of which are not
strictly synonyms, but hyponyms like ”Nu-Seals 300” or ”Bayer Extra Strength”
(ChEMBL via http://www.w3.org/2004/02/skos/core#altLabel). Includ-
ing all those alternative names in our search will result in a fairly complete
result, but may also be very inefficient.
There is also another risk involved. Elaborating on the previous example,
one of the alternative names is ASA. Although this is used as an alternative
name for aspirin, it is also the abbreviation of anti-sarcolemmal autoantibodies
Mus Musculus gene, of the disease Argininosuccinic aciduria and many more.
A search for this word will inevitably introduce many false positives. Another
source of ambiguity may be hypernymy, synonyms that are broader then the
actual submitted keyword.
2 Related work
While query expansion is a much studied subject we apply it here in highly spe-
cialized field of bio-medical sciences. This makes the use of tools like WordNet
as is done in [1, 2] less effective. However contrary to general language queries,
we have the advantage that the biomedical field is covered excellently by sev-
eral ontologies. Combining these multiple semantic ontologies in a very simple
heuristic to obtain an optimal query expansion, makes our approach distinct
from previous approaches.
3 The solution
We have opted to address the issues raised above with some simple heuristics,
which rely heavily on the fact that semantic platforms like DISQOVER bring
together multiple data sources. While each data source separately does not have
adequate power to discriminate, together they do.
We use the following algorithm to prune the synonym list
– Retrieve documents We retrieve all the documents that match the query
string exactly either with their preferred label or with one of the alternative
names. The provenance of each of these labels or names is also retrieved
– Merge documents the documents are merged if there is sufficient overlap
between their names and if the classes of the documents are compatible. As
an example consider a broad term like lung cancer. Multiple disease instances
match this term. By merging these instances and their synonyms we cover
the landscape the user probably wants to investigate
– Score synonyms Synonyms are scored based on the number of data set
that support them
#data sets − 1
score = if #data sets > 1
max(data sets) − 1
3
If data sets == 1 then the score is set 1. This scales the score between 0 and
1.
If a synonym is sufficiently short (currently 8 characters) then it is checked
whether it gives rise to false positives. This happens by retrieving the classes
of the documents having exact matches for that synonym. If more classes are
found than the current concept possesses, it gets a negative score for being
ambiguous. If a synonym is a number with less than 5 digits or less than 3
alphanumeric characters, it also gets a negative score for the same reason.
– Remove containing synonyms If a synonym contains another shorter
synonym, there is no reason to put it in a query. If the shorter word has a
lower score than the containing word, it inherits the highest score
We only retain the synonyms which have a score larger than 0.
In the example of aspirin we retain 12 possible synonyms (16%), the first
three being aspirin, acetylsalicylic acid and salicylic acid acetate.
4 Evaluation and results
We have analyzed the results of the 500 most prevalent keywords which were
recorded in DISQOVER in the first 6 months of 2017. For each keyword we have
ordered all the synonyms by score and by descending word length in case of a
tie for score . As an evaluation we submitted all these keywords to the search
engine in this order instead of submitting only the optimal synonyms and we
recorded the following per synonym
– How much hits we found in total by accumulating the synonyms with OR
– How much hits we get by only submitting the current synonym
An example of this output is shown in Appendix A. Per merged document we
register
– The optimal number of synonyms. We do this by re-ordering the synonyms
by descending number of hits they add and count how much synonyms would
we minimally need to obtain 98% of the hits
– Total number of synonyms per merged document
– The number of synonyms with score > 0
This enables us to measure what we miss if we only submit synonyms with a
score ≥ 0 .
We can split the results in two big groups, which we analyze separately. On
one hand we have keywords for which the score is not able to discriminate: so all
synonyms need to be checked. Basically this happens when there is only one data
source contributing to the concept or that all the data sources completely agree
on all alternative names. On the other hand we have the group for which the
method does make some difference and it does allow us to skip some synonyms.
The first group on average has 6.16 synonyms per keywords, while the second
group has 30.60 synonyms per keyword. In other words, in the cases the method
does not make any difference, there was not really a need for it to begin with.
4
Group Average % submitted Average Median
#synonyms missed hits missed hits
All synonyms considered 6.16 100% 0 0
Filtered by method 30.6 15.8% 9.4% 1.7%
In the second group we submit only 15.8% of all the synonyms if we apply
the method. We observe that the time to run the factor roughly scales with the
amount of synonyms we submit. If we compare the number of hits we obtain
by this subset to the hits by submitting all the synonyms, on average we miss
9.4%. However the median of the fraction of missed hits is 1.7%. So we have
a few very high outliers. What is causing this? The worst is example is DNM2
where we miss 94% of all the hits by only considering the synonyms with score
> 0. This is almost exclusively caused by one alternative name for this gene i.e.
Cytoskeletal protein (through UMLS). Although this gene is indeed related to
this concept, this concept is much broader. So it is a hypernymy of the submitted
concept and excluding it is actually beneficial: it would have given 16 times
more false positives if it were included. This is a pattern for most high miss
fraction examples. Consider atezolizumab. Here the hypernymy anti-pd-l1 (from
ChEMBL) is successfully excluded by our method. So we consider it justified
to exclude the high end misses tail and focus on the median: more than 98% of
the hits were found by less than 16% of the synonyms. The minimum number
of synonyms needed to get 98% of the hits is actually 9,01%, meaning that we
submit less than double of this absolute minimum.
For ambiguous synonyms we conducted a manual check for false positives
in a small subset of 20 clinical studies prominently containing ASA. Nine out
of twenty are not about aspirin at all: we found 3 about 5-aminosalicylates, 3
about the ASA-PS classification and 1 about resp. advanced surface ablation,
Avonex- Steroid Azathioprine and Argininosuccinic Aciduria. Of the other 11
only one does not contain one of the other synonyms included by our method.
So we avoid 45% false positives and trade these for 9% false negatives.
5 Conclusion
We have reduced the number of submitted synonyms by 84%, thereby losing only
1.7% of the hits. The false positive exclusion, which is a lot harder to check, also
seems to work well based on the small manually curated sample.
On a broader level we see that the actual number of synonyms needed to
attain most of the text hits is even lower: on average we only between 2 and 3
synonyms, the median is even 1! So for public data set it might be a hint to focus
more on quality then on quantity when choosing alternative names for concepts.
Overall we can conclude that the methods works well, despite the fact that
it is very simple. It is clearly a demonstration of the cheap gains we get by
combining multiple data sets in one semantic framework.
5
A APPENDIX Complete example: Aspirin
The example output we generated when evaluating Aspirin is presented here.
The synonyms are submitted in order of the table 1, which also contains the
results for each synonym. As you can see the first 2 synonyms bring the bulk
of the hits. The second one (Acetylsalicylic acid ) has about 8 times less hits
then Aspirin. But of those 10000 hits only about third is unique: the other
already have a hit for Aspirin. This pattern returns although all subsequent
synonyms return even less hits. The only synonym returning a significant number
of synonyms is ASA, but as mentioned in the text this is a very ambiguous word
with many different meanings. Many of these hits are identified as false positives.
Overall Aspirin has 75 alternative names. In the table we omit the containing
synonyms (such as aspirin sodium). In total 17.3% of the synonyms are submit-
ted (13) and we miss 1.04% of the hits. In the optimal case only 2.7% of the
synonyms have to be submitted to obtain 98% of the hits. The timings for the
queries are: 70 ms for retrieving hits for Aspirin only, 190 ms for retrieving all
hits for synonyms with score ≥ 0 and 2650 ms for all 75 synonyms.
synonym (1) (2) (3) (4) (5)
aspirin 1 100.00% 100.00% 84527 84526
acetylsalicylic acid 0.71 4.08% 12.19% 87974 10308
salicylic acid acetate 0.43 0.00% 0.01% 87977 13
2-(acetyloxy)benzoic acid 0.43 0.01% 0.24% 87989 208
acetylsalicylate 0.29 0.25% 0.81% 88210 709
2-acetoxybenzoic acid 0.29 0.01% 0.02% 88218 21
o-acetylsalicylic acid 0.29 0.00% 0.02% 88218 19
o-acetoxybenzoic acid 0.14 0.00% 0.00% 88218 2
2-acetyloxybenzoic acid 0.14 0.00% 0.01% 88221 8
o-carboxyphenyl acetate 0.14 0.00% 0.00% 88221 1
acidum acetylsalicylicum 0.14 0.00% 0.00% 88223 4
benzoic acid, 2-(acetyloxy)- 0.14 0.00% 0.02% 88227 19
2-acetoxybenzenecarboxylic acid 0.14 0.00% 0.00% 88227 1
levius 0 0.02% 0.04% 88245 31
platet 0 0.11% 0.13% 88344 111
asprin 0 0.08% 0.18% 88416 162
enprin 0 0.00% 0.00% 88416 1
asparin 0 0.00% 0.00% 88418 3
danamep 0 0.00% 0.00% 88418 1
paynocil 0 0.00% 0.01% 88419 12
Table 1. All synonyms is DISQOVER for aspirin. Columns (1) Score, (2) Percentage
of extra hits compared to all previous synonyms combined with OR ,(3) Percentage of
hits for this synonym only, (4) Number of total cumulative hits (all previous synonyms
combined with OR), (5) Number of hits for this synonym only
6
synonym (1) (2) (3) (4) (5)
measurin 0 0.00% 0.22% 88422 191
aspirine 0 0.10% 0.16% 88514 141
postmi 75 0 0.00% 0.00% 88514 3
gencardia 0 0.00% 0.00% 88514 1
aspro clr 0 0.00% 0.00% 88514 1
equi-prin 0 0.00% 0.00% 88514 1
disprin cv 0 0.00% 0.00% 88514 2
alka rapid 0 0.00% 0.00% 88514 1
postmi 300 0 0.00% 0.00% 88514 2
polopiryna 0 0.00% 0.01% 88518 7
acetophen 0 0.00% 0.00% 88518 2
angettes 75 0 0.00% 0.00% 88518 1
nu-seals 75 0 0.00% 0.00% 88518 2
nu-seals 300 0 0.00% 0.00% 88518 1
nu-seals 600 0 0.00% 0.00% 88518 1
8-hour bayer 0 0.00% 0.00% 88518 1
micropirin ec 0 0.00% 0.00% 88518 1
disprin direct 0 0.00% 0.00% 88518 1
acetosalic acid 0 0.00% 0.00% 88518 1
acetylsalic acid 0 0.00% 0.00% 88520 4
anadin all night 0 0.00% 0.01% 88521 12
2-acetoxybenzoate 0 0.05% 0.08% 88563 67
acetyl salicylate 0 0.02% 0.10% 88585 92
acetylsalicylsure 0 0.00% 0.02% 88589 19
nu-seals cardio 75 0 0.00% 0.00% 88589 1
azetylsalizylsure 0 0.00% 0.00% 88589 1
azetylsalizylsaeure 0 0.00% 0.00% 88589 1
acetylsalicylsaeure 0 0.00% 0.00% 88589 1
acetylsalisylic acid 0 0.00% 0.01% 88592 9
bayer extra strength 0 0.00% 0.00% 88592 2
acetylsalicyclic acid 0 0.29% 0.53% 88850 472
acetyl salicylic acid 0 0.47% 1.07% 89264 949
cido acetilsaliclico 0 0.03% 0.05% 89290 43
acetyl salicyclic acid 0 0.00% 0.02% 89294 14
2-acetoxy-benzoic acid 0 0.06% 0.08% 89348 69
acide actylsalicylique 0 0.00% 0.01% 89348 5
acetylsalicylicum acidum 0 0.00% 0.01% 89350 10
acide 2-(actyloxy)benzoque 0 0.00% 0.00% 89350 1
acide 2-(acetyloxy)benzoique 0 0.00% 0.00% 89350 1
(aspirin)2-acetoxy-benzoic acid 0 0.00% 0.00% 89350 2
2-(methoxycarbonyl)benzoic acid 0 0.00% 0.00% 89353 4
ecotrin 0.01% 0.25% 89359 227
asa 42.19% 48.64% 127053 43461
Table 2. The continuation of table 1
7
References
1. Voorhees, Ellen M. Query expansion using lexical-semantic relations. Proceedings of
the 17th annual international ACM SIGIR conference on Research and development
in information retrieval. Springer-Verlag New York, Inc., 1994.
2. Mandala, Rila, Takenobu Tokunaga, and Hozumi Tanaka. ”Combining multiple
evidence from different types of thesaurus for query expansion.” Proceedings of the
22nd annual international ACM SIGIR conference on Research and development in
information retrieval. ACM, 1999.