id	author	title	date	pages	extension	mime	words	sentence	flesch	summary	cache	txt
eblip-30014	Sharma, Ansh; Barrett, Kathryn; Stapelfeldt, Kirsta	Natural Language Processing for Virtual Reference Analysis	2022.0		.htm	text/html	6113	305	47	Consequently, our research team at the University of Toronto Scarborough (UTSC) Library sought to explore the viability of automatic and natural language processing (NLP) methodologies for chat transcript analysis. Ozeran and Martin (2019) also tested different algorithms for topic modeling using chat data from the University of Illinois at Urbana-Champaign, determining that Latent Dirichlet Allocation, Phrase-Latent Dirichlet Allocation, and Non-Negative Matrix Factorization had the most promise for large datasets.	cache/eblip-30014.htm	txt/eblip-30014.txt