1. An Automatic Construction of Arabic Similarity Thesaurus
Abdulaziz Al-Qabbany AbdulMalik Al-Salman Abdulrahman Almuhareb
CITALA 2009

2. Outline Introduction Thesauruses Similarity Thesaurus
Proposed Improvement
The Experiment
Evaluation
Discussion
Conclusions and Future Work

3. Introduction Thesaurus importance
Effective Information Retrieval systems
Vocabulary mismatch problem
Query Expansion

4. Thesauruses Arabic thesauruses Manual construction drawbacks:
cost
time
subjectivity
Automatic construction approaches

5. Similarity Thesaurus
Qiu and Frei (1993) presented their query expansion model using similarity thesaurus.
Zazo et al. (2005) used the same approach for constructing a Spanish similarity thesaurus.
Expanding queries based on similarity to their concepts rather than similarity to the individual terms.

6. Similarity Thesaurus (cont.)
Using similarity thesaurus is analogous to the translation from a language to another.
Example
قرص الدواء
قرص الشمس
قرص ضوئي
قـرص

7. Similarity Thesaurus construction
The similarity thesaurus is a matrix that represents terms similarities.
Each term is represented by a vector that determines its relation with each document.
The matrix is generated through calculating similarities between terms vectors.

8. Query Expansion using Similarity Thesaurus
Similarity between the query q and any term t is computed as the sum of the similarities values between each query term and t.
SIM_QT(q, t) =
As a response to any query, the terms can be ranked in descending order according to their SIM_QT values.

9. “Sum” method
“SUM” method is appropriate when the similarity values between the query terms and the indexed term are consistent within the same range.
When similarity values are inconsistent, the differences between the values will not be reflected on the total sum.
Similarity values are considered to be inconsistent when they contain outliers.

10. Outliers
Outlier is a value that is considerably dissimilar or inconsistent with the majority of the data.
outlier Y X

11. Proposed Improvement
A given term should have a high similarity value with each individual term in the query in order to be considered related.
The dispersion between the similarity values is one of the factors that needed to be considered in query expansion.
The total similarity value should remain as the main factor in query expansion.

12. Proposed Improvement (cont.)
Instead of using the sum of the similarity values, we use the mean of the values subtracted by the standard error of the mean (SE).
SIM_QT(q, t) =
The standard error of the mean is a measure of data dispersion.
SE =
where, α is the standard deviation and n is the number of values.

13. The Experiment
we used the France Press Agency Arabic news of years 2004, 2005 and 2006 as the document collection.
This document collection can be found in LDC Arabic Gigaword corpus (Third Edition).
After examining the high frequency terms in the collection, we had chosen 150 stop words.

14. Document collection characteristics
Number
Number of Documents
208,596
Number of Terms
435,846
Total Number of Terms Occurrences
30,415,222
Average Number of Words per Document
69.78
Number of Processed Terms
248,311

15. Evaluation
The objective of the evaluation was to assess the relevance strength of the produced terms.
The evaluation process was applied for both the “SUM” and “MEAN” methods.
We have selected twenty common topics that belong to five different domains.

16. Evaluation
For each topic, the top ten related terms were presented to five expert evaluators.
Each evaluator was asked to study these twenty topics carefully and then specify if the produced terms are relevant or not.
Levels of relevance:
Relevant
Somewhat Relevant
Irrelevant

17. Evaluation Results
The relevance strength of the standard “SUM” method was 95.0%, while the Relevance strength of the “MEAN” method was 98.1%.

18. Discussion
We believe that the main reason that makes the “MEAN” method a better method is its ability to detect and exclude outliers.
Adding a single term to the query may completely change the concept of the query.
The candidate related term should have consistent similarities with all of the query terms.

19. MEAN Most Related Terms
Example
The response to a query about the former French president “جاك شيراك”:
SUM Most Related Terms
Value
الفرنسي
0.814
الاليزيه
0.630
فرنسا
0.556
الرئيس
0.503
سترو
0.482
MEAN Most Related Terms
Value
الفرنسي
0.383
الاليزيه
0.283
فرنسا
0.266
الرئيس
0.241
باريس
0.214

20. Example (cont.)

21. Conclusions
The relevance strength of the standard “SUM” method was 95.0%, while the Relevance strength of the “MEAN” method was 98.1%
“MEAN” method shows an improvement of about 3.3% over “SUM” method.
We conclude that the “MEAN” method is more accurate mainly because it can detect and exclude the outliers.

22. Future Work Applying word stemming. Producing collocations.
Constructing a single word-category thesaurus.
Using similarity thesaurus in question answering.

23. End