KIT | KIT-Bibliothek | Impressum | Datenschutz

Programming in Natural Language with fuSE : Synthesizing Methods from Spoken Utterances Using Deep Natural Language Understanding

Weigelt, Sebastian; Steurer, Vanessa; Hey, Tobias ORCID iD icon; Tichy, Walter F. ORCID iD icon

Abstract (englisch):

The key to effortless end-user programming is natural language. We examine how to teach intelligent systems new functions, expressed in natural language. As a first step, we collected 3168 samples of teaching efforts in plain English. Then we built fuSE, a novel system that translates English function descriptions into code. Our approach is three-tiered and each task is evaluated separately. We first classify whether an intent to teach new functionality is present in the utterance (accuracy: 97.7% using BERT). Then we analyze the linguistic structure and construct a semantic model (accuracy: 97.6% using a BiLSTM). Finally, we synthesize the signature of the method, map the intermediate steps (instructions in the method body) to API calls and inject control structures (F1: 67.0% with information retrieval and knowledge-based methods). In an end-to-end evaluation on an unseen dataset fuSE synthesized 84.6% of the method signatures and 79.2% of the API calls correctly.

Verlagsausgabe §
DOI: 10.5445/IR/1000124407
Veröffentlicht am 09.10.2020
DOI: 10.18653/v1/2020.acl-main.395
Zitationen: 3
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Programmstrukturen und Datenorganisation (IPD)
Publikationstyp Proceedingsbeitrag
Publikationsjahr 2020
Sprache Englisch
Identifikator ISBN: 978-1-952148-25-5
KITopen-ID: 1000124407
Erschienen in Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Ed.: D. Jurafsky
Veranstaltung 58th Annual Meeting of the Association for Computational Linguistics (ACL 2020), Online, 05.07.2020 – 10.07.2020
Verlag Association for Computational Linguistics (ACL)
Seiten 4280–4295
Bemerkung zur Veröffentlichung Die Veranstaltung fand wegen der Corona-Pandemie als Online-Event statt
Nachgewiesen in Dimensions
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page