Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


41591.pdf (13.77 MB)

ETD Abstract Container

Abstract Header

A Systematic Comparative Study of Sentence Embedding Methods Using Real-World Text Corpora

Mistry, Deven Mahesh
http://orcid.org/0000-0001-5084-8112
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1637311155942699

Abstract Details

2021, MS, University of Cincinnati, Engineering and Applied Science: Computer Science.
Many natural language processing (NLP) tasks require the conversion of textual data to numeric representations. Vector-space representations are the most popular way to do this. Initial vector-space models were used to represent individual words, but several very complex language models have been developed recently that can generate vector-space representations of sentences, paragraphs, and even entire documents. These models use various deep learning architectures including simple RNNs, stacked LSTMs, and Transformers [54]. Typically, the models are evaluated on synthetic or carefully curated benchmark datasets such as GLUE [56], SQuAD [45], COCO [55], etc. and tasks such as sentiment analysis and text classification. However, it is often unclear whether performance on these controlled benchmarks can transfer to non-curated, real-world datasets with uncontrolled semantic noise and complex structure. The goals of this thesis are: 1) To develop a methodology for systematically comparing a representative set of sentence encoder models on real-world texts; and 2) To apply this methodology using several sizeable real-world texts to arrive at a definitive ranking of the methods. The methodology uses the pattern of semantic similarity between sentence pairs to obtain a representation of semantic structure for each document using each encoding method. These structures are then compared statistically, through visualization, and through manual scoring to assess the relative quality of the representations produced by each encoding method. An innovative aspect of this research is the use of multiple English language translations of the same text as a further cross-validation mechanism.
Ali Minai, Ph.D. (Committee Chair)
Anca Ralescu (Committee Member)
Raj Bhatnagar, Ph.D. (Committee Member)
110 p.
Computer Science

Recommended Citations

Citations

  • Mistry, D. M. (2021). A Systematic Comparative Study of Sentence Embedding Methods Using Real-World Text Corpora [Master's thesis, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1637311155942699

    APA Style (7th edition)

  • Mistry, Deven Mahesh. A Systematic Comparative Study of Sentence Embedding Methods Using Real-World Text Corpora. 2021. University of Cincinnati, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1637311155942699.

    MLA Style (8th edition)

  • Mistry, Deven Mahesh. "A Systematic Comparative Study of Sentence Embedding Methods Using Real-World Text Corpora." Master's thesis, University of Cincinnati, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1637311155942699

    Chicago Manual of Style (17th edition)

ucin1637311155942699
175
© 2021, some rights reserved.Creative Commons License A Systematic Comparative Study of Sentence Embedding Methods Using Real-World Text Corpora by Deven Mahesh Mistry is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License. Based on a work at etd.ohiolink.edu.
This open access ETD is published by University of Cincinnati and OhioLINK.