--- title: "A Study of the Effect of Resolving — AL-Khawaldeh 2019" description: "The paper addresses Arabic recognizing textual entailment (RTE)." authors: ["Fatima T. AL-Khawaldeh"] year: 2019 venue: "arXiv:1907.03871 · World of Computer Science and Information Technology Journal (WCSIT) ISSN: 2221-0741 Vol. 5, No. 7, 124-128, 2015" source: "https://arxiv.org/abs/1907.03871" doi: "10.48550/arxiv.1907.03871" arxiv: "1907.03871" licence: "CC BY 4.0 (arXiv)" licence_url: "http://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/al-2019-khawaldeh-study-effect-resolving-negation-sentiment/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/al-2019-khawaldeh-study-effect-resolving-negation-sentiment/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2019](https://md.research.elamur.ai/free/en/papers/2019/) / A Study of the Effect of Resolving Negation and Sentiment Analysis in Recognizing Text Entailment for Arabic # A Study of the Effect of Resolving Negation and Sentiment Analysis in Recognizing Text Entailment for Arabic **Authors:** [Fatima T. AL-Khawaldeh](https://md.research.elamur.ai/free/en/authors/fatima-t-al-khawaldeh/) · **Year:** 2019 · **Venue:** arXiv:1907.03871 · World of Computer Science and Information Technology Journal (WCSIT) ISSN: 2221-0741 Vol. 5, No. 7, 124-128, 2015 · **Licence:** [CC BY 4.0 (arXiv)](http://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://arxiv.org/abs/1907.03871) · [PDF](https://arxiv.org/pdf/1907.03871) · [DOI 10.48550/arxiv.1907.03871](https://doi.org/10.48550/arxiv.1907.03871) · [arXiv 1907.03871](https://arxiv.org/abs/1907.03871) · [OpenAlex](https://openalex.org/W2959628155) **Citations:** 12 citations (openalex-snapshot · 2026-10-14) · 1 in our library, of the 140884 papers whose bibliographies we read ## Summary The paper addresses Arabic recognizing textual entailment (RTE). An earlier lexical system, called ATE here, accepted pairs it should have rejected for two reasons: it discarded negation particles as stop words, and it ignored whether text and hypothesis express opposite opinions. The author adds a layer on top of ATE, called SANATE. ATE works in four steps: - it stems words; - it expands them with Arabic WordNet senses; - it counts shared words; - it tests three cosine-style overlap scores against fixed thresholds. SANATE acts only when ATE outputs "entails," and it adds two checks: - **Negation rules.** These test whether a negative particle precedes a shared verb on one side, on both sides, or precedes unshared verbs. - **Polarity check.** A lexicon-based classifier, following the CNSAMSA-SAT tool and using multi-domain lexicons, labels each side positive, negative or neutral. Mismatched polarity blocks entailment. On ArbTEDS (618 pairs), accuracy rises from 0.617 to 0.693. The work matters as an early Arabic demonstration that cheap, interpretable negation and polarity checks fix systematic errors of overlap-based entailment. ## Key points - The baseline ATE computes cosT = √(c/m), cosH = √(c/n) and a joint score. Here c is the number of shared words (same root or WordNet-related), m is the text length and n is the hypothesis length. Its thresholds are τ1=0.095, τ2=0.2 and τ3=0.5. - Five Arabic negative particles are handled: maa, laa, lam, lan and laysa. In the baseline, laa was simply deleted as a stop word. - The negation rules work as follows: - a shared verb negated on only one side means not entail; - a shared verb negated on both sides means entail, even when the verbs are WordNet-related rather than identical; - negation on non-shared verbs means not entail; - with several shared verbs, the same logic is applied to each verb. - The polarity component uses ALL_lex.csv, a combined lexicon from ElSahar & El-Beltagy covering hotel, library, movie, product and restaurant domains. It assigns a label by counting positive and negative lexicon hits, and a positive or negative label needs at least 2 hits plus a majority. - Evaluated on ArbTEDS (618 T–H pairs), accuracy is 0.617 for ATE and 0.693 for SANATE. ## Interesting ideas - A second-stage checker runs only on positive "entails" decisions from the overlap model, acting mainly as a veto, so it cannot damage pairs already rejected. — For Mova (answer): A yes/entailed answer could pass through a separate contradiction-veto pass over the world state, keeping the base reasoner simple and each veto traceable. - Negation is compared per aligned shared predicate, not per sentence. Negation on both sides cancels out, while negation on one side breaks entailment. — For Mova (frames): Store polarity as an attribute on each frame, then compare it only between frames aligned by VerbNet class or lemma, so double negation across T and H is handled correctly. - The paper's core failure diagnosis is that stop-word removal silently erased a truth-flipping word, a cautionary design lesson for any pipeline that filters tokens. — For Mova (tokens): Guarantee that negators and Polarity=Neg features always survive into lemmas and the UD tree, and add a regression test that fails if any is dropped. - Matching opinion polarity is treated as a cheap necessary condition: an evaluatively positive statement cannot entail a negative one. — For Mova (world): Record evaluative stance of entities and events in the text's world, and refuse entailment or answers whose stance contradicts the stored one. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* Fields: [Linguistics & languages](https://md.research.elamur.ai/free/en/shelves/linguistics/) ## Cite The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @misc{alkhawaldeh2019study, title = {{A Study of the Effect of Resolving Negation and Sentiment Analysis in Recognizing Text Entailment for Arabic}}, author = {AL-Khawaldeh, Fatima T.}, year = {2019}, doi = {10.48550/arxiv.1907.03871}, eprint = {1907.03871}, archivePrefix = {arXiv}, url = {https://doi.org/10.48550/arxiv.1907.03871} } ``` - [← previous A Structural Probe for Finding Syntax in Word Representations](https://md.research.elamur.ai/free/en/papers/hewitt-2019-structural-probe-finding-syntax-word/) - [next → A Survey on Bias and Fairness in Machine Learning](https://md.research.elamur.ai/free/en/papers/mehrabi-2019-survey-bias-fairness-machine-learning/) - [↑ back to papers of 2019](https://md.research.elamur.ai/free/en/papers/2019/) ## References 15 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Swapnil G. and Bhattacharya A., "Survey in Textual Entailment," Center for Indian Language Technology, 2014. [find](https://scholar.google.com/scholar?q=Swapnil+G%2E+and+Bhattacharya+A%2E%2C+%22Survey+in+Textual+Entailment%2C%22+Center+for+Indian+Language+Technology%2C+2014%2E) 2. Dagan I., Glickman O., and Magnini B.," The pascal recognizing textual entailment challenge," In Proceedings of the second PASCAL challenges workshop on recognizing textual entailment, 2005. [find](https://scholar.google.com/scholar?q=Dagan+I%2E%2C+Glickman+O%2E%2C+and+Magnini+B%2E%2C%22+The+pascal+recognizing+textual+entailment+challenge%2C%22+In+Proceedings+of+the+second+PASCAL+challenges+workshop+on+recognizing+textual+entailment%2C+2005%2E) 3. Hickl A., Williams J., Bensley J., and Roberts K., "Recognizing textual entailment with LCC's GROUNDHOG system," In Proceedings of the 26 Recognizing Textual Entailment Challenge, 2006. [find](https://scholar.google.com/scholar?q=Hickl+A%2E%2C+Williams+J%2E%2C+Bensley+J%2E%2C+and+Roberts+K%2E%2C+%22Recognizing+textual+entailment+with+LCC%27s+GROUNDHOG+system%2C%22+In+Proceedings+of+the+26+Recognizing+Textual+Entailment+Challenge%2C+2006%2E) 4. Hickl A. and Bensley J.," A discourse commitment-based framework for recognizing textual entailment," In Proceedings of the ACL-PASCAL Workshop on Textual Entailment and Paraphrasing - RTE '07, Morristown, NJ, USA,. Association for Computational Linguistics, 2007. [find](https://scholar.google.com/scholar?q=Hickl+A%2E+and+Bensley+J%2E%2C%22+A+discourse+commitment%2Dbased+framework+for+recognizing+textual+entailment%2C%22+In+Proceedings+of+the+ACL%2DPASCAL+Workshop+on+Textual+Entailment+and+Paraphrasing+%2D+RTE+%2707%2C+Morris) 5. Lapponi E., Read J., and Øvrelid L., "Representing and Resolving Negation for Sentiment Analysis," In Proceedings of the 2012 ICDM Workshop on Sentiment Elicitation from Natural Text for Information Retrieval and Extraction. Brussels, Belgium, 2012. [find](https://scholar.google.com/scholar?q=Lapponi+E%2E%2C+Read+J%2E%2C+and+%C3%98vrelid+L%2E%2C+%22Representing+and+Resolving+Negation+for+Sentiment+Analysis%2C%22+In+Proceedings+of+the+2012+ICDM+Workshop+on+Sentiment+Elicitation+from+Natural+Text+for+Information+R) 6. Alabbas M., "ArbTE: Arabic Textual Entailment," In Proceedings of the Student Research Workshop associated with RANLP 2011, Hissar, Bulgaria, 2011. [find](https://scholar.google.com/scholar?q=Alabbas+M%2E%2C+%22ArbTE%3A+Arabic+Textual+Entailment%2C%22+In+Proceedings+of+the+Student+Research+Workshop+associated+with+RANLP+2011%2C+Hissar%2C+Bulgaria%2C+2011%2E) 7. Zhang K. and Shasha D., "Simple fast algorithms for the editing distance between trees and related problems," SIAM J. Computer., vol. 18, no.6, pp.1245–1262, 1989. [find](https://scholar.google.com/scholar?q=Zhang+K%2E+and+Shasha+D%2E%2C+%22Simple+fast+algorithms+for+the+editing+distance+between+trees+and+related+problems%2C%22+SIAM+J%2E+Computer%2E%2C+vol%2E+18%2C+no%2E6%2C+pp%2E1245%E2%80%931262%2C+1989%2E) 8. Alabbas M., "A Dataset for Arabic Textual Entailment," In Proceedings of the Student Research Workshop associated with RANLP 2013, pp. 7–13, Hissar, Bulgaria, 2013. [find](https://scholar.google.com/scholar?q=Alabbas+M%2E%2C+%22A+Dataset+for+Arabic+Textual+Entailment%2C%22+In+Proceedings+of+the+Student+Research+Workshop+associated+with+RANLP+2013%2C+pp%2E+7%E2%80%9313%2C+Hissar%2C+Bulgaria%2C+2013%2E) 9. Tatar D., Mihis A., and Lupsa D., "Entailment–based Linear Segmentation in Summarization", International Journal of Software Engineering and Knowledge Engineering vol. 19, no. 80, pp. 1023–1038, 2009. [find](https://scholar.google.com/scholar?q=Tatar+D%2E%2C+Mihis+A%2E%2C+and+Lupsa+D%2E%2C+%22Entailment%E2%80%93based+Linear+Segmentation+in+Summarization%22%2C+International+Journal+of+Software+Engineering+and+Knowledge+Engineering+vol%2E+19%2C+no%2E+80%2C+pp%2E+1023%E2%80%931038%2C+2009%2E) 10. AL-Khawaldeh F. and Samawi V., "Lexical Cohesion and Entailment based Segmentation for Arabic Text Summarization (LCEAS)," The World of Computer Science and Information Technology Journal (WSCIT), vol. 5, no. 3, pp. 51-60, 2015 [find](https://scholar.google.com/scholar?q=AL%2DKhawaldeh+F%2E+and+Samawi+V%2E%2C+%22Lexical+Cohesion+and+Entailment+based+Segmentation+for+Arabic+Text+Summarization+%28LCEAS%29%2C%22+The+World+of+Computer+Science+and+Information+Technology+Journal+%28WSCIT%29%2C+vol) 11. AL-Khawaldeh F., "Answer Extraction for Why Arabic Questions Answering Systems: EWAQ," The World of Computer Science and Information Technology Journal (WSCIT), vol.5, no. 5, pp. 82.86, 2015 [find](https://scholar.google.com/scholar?q=AL%2DKhawaldeh+F%2E%2C+%22Answer+Extraction+for+Why+Arabic+Questions+Answering+Systems%3A+EWAQ%2C%22+The+World+of+Computer+Science+and+Information+Technology+Journal+%28WSCIT%29%2C+vol%2E5%2C+no%2E+5%2C+pp%2E+82%2E86%2C+2015) [our card](https://md.research.elamur.ai/free/en/papers/al-2019-khawaldeh-answer-extraction-why-arabic-questions/) 12. Abbasi A., Chen H., and Salem A, "Sentiment analysis in multiple languages: Feature selection for opinion classification in Web forums," ACM Transactions on Information Systems (TOIS), vol. 26, no.3, 2008. [find](https://scholar.google.com/scholar?q=Abbasi+A%2E%2C+Chen+H%2E%2C+and+Salem+A%2C+%22Sentiment+analysis+in+multiple+languages%3A+Feature+selection+for+opinion+classification+in+Web+forums%2C%22+ACM+Transactions+on+Information+Systems+%28TOIS%29%2C+vol%2E+26%2C+no%2E3%2C+) 13. Ahmad K., Cheng D., and Almas Y.," Multi-lingual sentiment analysis of financial news Streams," In Proceedings of the 1st International Conference on Grid in Finance, 2006. [find](https://scholar.google.com/scholar?q=Ahmad+K%2E%2C+Cheng+D%2E%2C+and+Almas+Y%2E%2C%22+Multi%2Dlingual+sentiment+analysis+of+financial+news+Streams%2C%22+In+Proceedings+of+the+1st+International+Conference+on+Grid+in+Finance%2C+2006%2E) 14. Elhawary M. and Elfeky M., "Mining Arabic Business Reviews," In Proceedings of International Conference on Data Mining Workshops (ICDMW), pp. 1108–1113, 2010. [find](https://scholar.google.com/scholar?q=Elhawary+M%2E+and+Elfeky+M%2E%2C+%22Mining+Arabic+Business+Reviews%2C%22+In+Proceedings+of+International+Conference+on+Data+Mining+Workshops+%28ICDMW%29%2C+pp%2E+1108%E2%80%931113%2C+2010%2E) 15. El-Halees A., "Arabic Opinion Mining Using Combined Classification Approach," In Proceedings of the International Arab Conference on Information Technology (ACIT), 2011. [find](https://scholar.google.com/scholar?q=El%2DHalees+A%2E%2C+%22Arabic+Opinion+Mining+Using+Combined+Classification+Approach%2C%22+In+Proceedings+of+the+International+Arab+Conference+on+Information+Technology+%28ACIT%29%2C+2011%2E)