Open AccessVECTOR SPACE MODELS OF KYIV CITY PETITIONS
Author(s) -
Roman Shaptala,
Gennadiy Kyselov
Publication year - 2021
Publication title -
information, computing and intelligent systems
Language(s) - English
Resource type - Journals
eISSN - 2708-4930
pISSN - 2786-8729
DOI - 10.20535/2708-4930.2.2021.244188
Subject(s) - word2vec , space (punctuation) , vector space model , vector space , word (group theory) , curse of dimensionality , computer science , vector (molecular biology) , dimensionality reduction , information retrieval , artificial intelligence , data mining , natural language processing , mathematics , biochemistry , chemistry , geometry , gene , recombinant dna , operating system
In this study, we explore and compare two ways of vector space model creation for Kyiv city petitions. Both models are built on top of word vectors based on the distributional hypothesis, namely Word2Vec and FastText. We train word vectors on the dataset of Kyiv city petitions, preprocess the documents, and apply averaging to create petition vectors. Visualizations of the vector spaces after dimensionality reduction via UMAP are demonstrated in an attempt to show their overall structure. We show that the resulting models can be used to effectively query semantically related petitions as well as search for clusters of related petitions. The advantages and disadvantages of both models are analyzed.