A Morphological, Syntactic and Semantic Search Engine for Hebrew Texts

What it is about

In unpointed Hebrew a string has on average about three readings, and a plain string search returns mostly wrong hits. Ornan's engine works in three stages. First the text is transliterated into a phonemic Latin script (ISO 259-3), where vowels and attached particles are explicit. A morphological analyzer then lists every reading of each string. The syntactic stage combines readings across the words of one clause, and a semantic stage checks them against a verb-centred conceptual dictionary: each verb entry lists its noun-phrase slots, their thematic roles, the semantic features and the governing prepositions, with about 10,000 verb entries and 35,000 noun entries. The author reports that about 95% of results match the query. The evaluation is described only in summary.

Ideas

• I1. A phonemic pre-script before analysis: map each unpointed token to a vowelled, particle-split form first, so that the splits are visible and the analyzer sees one reading per string. In tools/hbo this would sit between tokens and the segmenter, and it would make the pre and lex tables easier to audit.
• I2. Verb case frames as constraints: for each verb, list the required argument slots with their prepositions and feature constraints, and use them to filter the parses that gparse proposes. The paper's entries (for instance a verb of riding that takes a preposition for the animal, another for the goal, and another for the source) show the format.
• I3. A counting test: the paper's example sentence gives 2×3×8×3 = 144 candidate combinations before syntactic filtering. Run the same count on the tools/hbo readings of a fixed sentence, before and after the filter, as a measure of how much the filter removes.

For Mova

I1 goes into tools/hbo and 1. Script and pointing. I2 goes into Syntax: the clause and the sentence, with the frames in The lexicon. I3 is a test set for tools/hbo.

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.3115/1118637.1118645

@inproceedings{ornan2002morphological,
  title = {{A Morphological, Syntactic and Semantic Search Engine for Hebrew Texts}},
  author = {Ornan, Uzzi},
  booktitle = {Proceedings of the ACL-02 Workshop on Computational Approaches to Semitic Languages},
  year = {2002},
  doi = {10.3115/1118637.1118645},
  url = {https://doi.org/10.3115/1118637.1118645}
}

References

23 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Bentor,E., A.Angel, D.Ben-Ari-Segev and A.Lavie. 1992. Computerized Analysis of Hebrew Words in Hebrew Computational Linguistics, ed. by U. Ornan, G.Arieli and E.Doron, Ministry of Science and Technology, pages 36-38.(Hebrew). find
2. Carmel, David and Yoelle Maarek. 1999. Morphological disambiguation for Hebrew. In Proceedings of the 4 th International Workshop NGIT-99, Lecture notes in computer science 1649. Springer Verlag, pages 312-325. find
4. Chomsky, N. 1965. Aspects of the Theory of Syntax. MIT Press. find
5. Chomsky, N. Lectures on Government and Binding . . . Foris Pub. Choueka, Y. and Serge Lusignan.1985 . Disambiguation by Short Contexts. In Computers And Humanities Vol 19:147-157. find
6. Choueka Yaacov.1990. Responsa: An Operational Full-Text Retrieval System. In Computers in Literary and Linguistic Research. edited by J .Hamesse and A. Zampoli. Champion-Slatkine Paris-Geneve . Pages 94-102 find
7. Even-Shoshan, Avraham.1987. The New Dictionary. (Hebrew). find
8. Fillmore,C.C.1968. The Case for Case. In Universals in Linguistic Theory . Edited By E.Bach and R.Harmes. Holt, Rinehart and Winston, New York, Academic Press. Pages 59-81. find
9. Herz,Y. and M.Rimon.1992. Diminishing Ambiguity by Short-Context Automaton. In Hebrew Computational Linguistics . Edited by U. Ornan, G.Arieli and E. Doron. Ministry of Science and Technology. Pages 74-87.(Hebrew). find
10. Ide, Nancy and Jean Véronis.1998. Introduction to the Special Issue on Word Sense Disambiguation: The State of the Art. Compuational Linguistics 24:140 . find
11. ISO 259-3.1999 .Conversion of Hebrew Characters Into Latin Characters. Part 3: Phonemic Conversion. ISO/TC46/SC2. find
12. Levinger, Moshe. 1992. Morphological Disambiguation in Hebrew. Research Thesis for MSc in Computer Science. Technion. Haifa (Hebrew). find
13. Levinger,Moshe, Uzzi Ornan and Itai Alon. 1995. Learning Morpho-Lexical Probabilities from an Untagged Corpus with an Application to Hebrew. Computational Linguistics 21:383404 . find our card
14. Miller, George A.1993. Nouns in WordNet. Web file. find
15. Nirenburg, Sergei and Y. Ben Asher. 1984. HUHU :Hebrew University Hebrew Understander. Journal of Computational Linguistics Vol.9:161-182. find
16. Ornan, Uzzi. 1987. Hebrew Text Processing Based on Unambiguous Script. .Mishpatim 17:15-24.(Hebrew) find
17. Ornan, Uzzi.1991. Theoretical Gemination in Israeli Hebrew. Semitic Studies in honor of Wolf Lwslau. Edited by Alan S, Kaye. Otto Harrassowitz, Weisbaden. Pages 1158-1168. find
18. Ornan, Uzzi and Michael Katz. 1994. A New Program for Hebrew Index Based on the Phonemic Script .TR #LCL 94-7 (revised). Technion - I.I.T. find
19. Ostler, Nicholas. 1995. Perception Vocabulary in five Languages – Towards an Analysis Using Frame Elements. In Steffens Petra (editor) Machine Translation and the Lexicon. Springer Verlag. Pages 219-23. find
20. Segal, Erel,1999. Hebrew Morphological Analyzer for Hebrew undotted Analysis. Thesis for MSc in Computer Science. Technion. Haifa (Hebrew). find
21. Somers, H.L. 1987.Valency and Case in . . find
22. Stern, Naftali. 1994. The Verb Dictionary. Bar-Ilan University. (Hebrew). find
23. Wintner, Shuly and Uzzi Ornan. 1995. Syntactic Analysis of Hebrew Sentence. Natural Lnaguage Engineering 1:261-288. find
24. Whorf, Benjamin Lee. 1956. The Relation of Habitual Thought and Behavior to Language. In Leslie Spier (editor) Language, Culture and Reality , essays in memory of Edward Sapir.1941. Pages 75-93. Reprinted in John B. Carrol (editor) Language, Thought and Reality. M.I.T. Press. Pages 134-159 find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/aclbulk-2002-ornan-morphological-syntactic-semantic-search-engine/