Li, Teng: Interpreting Text Classification with Human-Understandable Counterfactual Instances. Hannover : Gottfried Wilhelm Leibniz Universität, Master Thesis, 2022, 26 S. DOI: http://doi.org/10.15488/11892
Zusammenfassung: | |
As the omnipresent machine learning models play increasingly important roles in our society, powerful interpretation tools to uncover their black boxes are needed.On the other hand, proven by psychological study, we humans are more likely to learn new concepts presented with contrastive instances.Therefore, interpreting ML models using the contrast between the original data instance and its counterfactuals has become a popular problem.Traditional counterfactual interpretation approaches tend to generate counterfactuals faithful to the ML model.However, they have little or no constraint on the meaningfulness of generated counterfactuals.This thesis proposes an approach generating a meaningful counterfactual interpretation of text classification models constrained with cosine similarity and POS (part-of-speech) properties of tokens.In this thesis, I use the text CNN model based on Kims Cnn\cite{KimsCnn} with fine-tuned Word2Vec embedding layer as the model to interpret.Then for the counterfactual generation, I leverage token-level HotFlip\cite{hotflip} and replace tokens under several constraints.Lastly, I will present that my approach results in more meaningful counterfactual interpretations compared with the vanilla HotFlip approaches using several examples. | |
Lizenzbestimmungen: | CC BY 3.0 DE |
Publikationstyp: | MasterThesis |
Publikationsstatus: | publishedVersion |
Erstveröffentlichung: | 2022 |
Die Publikation erscheint in Sammlung(en): | Fakultät für Elektrotechnik und Informatik |
Pos. | Land | Downloads | ||
---|---|---|---|---|
Anzahl | Proz. | |||
1 | Germany | 96 | 46,15% | |
2 | United States | 33 | 15,87% | |
3 | China | 10 | 4,81% | |
4 | Canada | 6 | 2,88% | |
5 | India | 5 | 2,40% | |
6 | France | 5 | 2,40% | |
7 | No geo information available | 4 | 1,92% | |
8 | Morocco | 4 | 1,92% | |
9 | Hong Kong | 4 | 1,92% | |
10 | United Kingdom | 3 | 1,44% | |
andere | 38 | 18,27% |
Hinweis
Zur Erhebung der Downloadstatistiken kommen entsprechend dem „COUNTER Code of Practice for e-Resources“ international anerkannte Regeln und Normen zur Anwendung. COUNTER ist eine internationale Non-Profit-Organisation, in der Bibliotheksverbände, Datenbankanbieter und Verlage gemeinsam an Standards zur Erhebung, Speicherung und Verarbeitung von Nutzungsdaten elektronischer Ressourcen arbeiten, welche so Objektivität und Vergleichbarkeit gewährleisten sollen. Es werden hierbei ausschließlich Zugriffe auf die entsprechenden Volltexte ausgewertet, keine Aufrufe der Website an sich.