2026-09-19T12:17:50Zhttps://dspace.mit.edu/server/oai/request

oai:dspace.mit.edu:1721.1/1449012022-08-30T03:03:05Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023

Glass, James Wagner, Julia N. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science 2022-08-29T16:19:44Z 2022-08-29T16:19:44Z 2022-05 2022-06-30T15:52:18.667Z https://hdl.handle.net/1721.1/144901 The natural language processing field has seen task-oriented dialog systems emerge as a strong area of interest in research and industry over the past years. However, the limited existence of complex and sufficiently annotated training data still places a bottleneck on the development of more advanced, domain-agnostic chatbots. Novel domains require extensive time and manual effort from experts when creating intents for new datasets to support dialog systems. This thesis analyzes a two-staged unsupervised semantic clustering and intent generation approach with multiple dataset adaptive interchangeable methods. We examine various pre-trained embeddings, scoring objectives for the number of clusters, unsupervised clustering algorithms, intent generation techniques, and utterance tokenization schemes. We then run experiments with these combinations on three datasets: SNIPS, MultiWOZ, and real-world chat data. This is followed by quantitative metric and in-depth qualitative cluster-based evaluation. We show the benefits of bigram frequency intent generation as datasets increase irregularity and confirm the success of the universal sentence encoder embeddings with K-Means clustering. Additionally, our examination of real-world data underlines the importance of fine-grained utterance tokenization and gives promise to the feasibility of research methods on unpublished data. Altogether, this thesis provides a comprehensive analysis covering the abilities of the the two-stage pipeline components to support open intent discovery for a variety of dataset characteristics, offering alternative solutions where beneficial for real-world applications. This gives insight to the optimal configuration to automatically generate a novel dialog training dataset from unstructured, unlabeled chat utterances. The code for this thesis can be found at https://github.com/jnwagner53/dialog-intent-generation. M.Eng. Massachusetts Institute of Technology In Copyright - Educational Use Permitted Copyright MIT http://rightsstatements.org/page/InC-EDU/1.0/ Open Intent Generation Through Unsupervised Semantic Clustering of Task-Oriented Dialog Thesis application/pdf Master Master of Engineering in Electrical Engineering and Computer Science Publication unknown unknown <Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="ed9af36a-bde7-4f62-93c7-24d4fe69be7e"> <Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type> <Title>Open Intent Generation Through Unsupervised Semantic Clustering of Task-Oriented Dialog</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2022-05</PublicationDate> <Authors> <Author> <DisplayName>Wagner, Julia N.</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <DisplayName>Massachusetts Institute of Technology</DisplayName> <OrgUnit /> </Publisher> </Publishers> <License>http://rightsstatements.org/page/InC-EDU/1.0/</License> <Abstract>The natural language processing field has seen task-oriented dialog systems emerge as a strong area of interest in research and industry over the past years. However, the limited existence of complex and sufficiently annotated training data still places a bottleneck on the development of more advanced, domain-agnostic chatbots. Novel domains require extensive time and manual effort from experts when creating intents for new datasets to support dialog systems. This thesis analyzes a two-staged unsupervised semantic clustering and intent generation approach with multiple dataset adaptive interchangeable methods. We examine various pre-trained embeddings, scoring objectives for the number of clusters, unsupervised clustering algorithms, intent generation techniques, and utterance tokenization schemes. We then run experiments with these combinations on three datasets: SNIPS, MultiWOZ, and real-world chat data. This is followed by quantitative metric and in-depth qualitative cluster-based evaluation. We show the benefits of bigram frequency intent generation as datasets increase irregularity and confirm the success of the universal sentence encoder embeddings with K-Means clustering. Additionally, our examination of real-world data underlines the importance of fine-grained utterance tokenization and gives promise to the feasibility of research methods on unpublished data. Altogether, this thesis provides a comprehensive analysis covering the abilities of the the two-stage pipeline components to support open intent discovery for a variety of dataset characteristics, offering alternative solutions where beneficial for real-world applications. This gives insight to the optimal configuration to automatically generate a novel dialog training dataset from unstructured, unlabeled chat utterances. The code for this thesis can be found at https://github.com/jnwagner53/dialog-intent-generation.</Abstract> <Access xmlns="http://purl.org/coar/access_right" > </Access> </Publication>