other · Figshare · 2018
Query Expansion
Rahmi, Adel
Cited by 0Open access
Abstract
Methods:<br>Query expansion uses several processes from Natural Language Processing. In most cases when implementing query understanding systems, a simple or sometimes very complex spell checker is used. In this project; however, our first step in this project is normalization. In normalization process we use Hazm library normalizer. The output undergoes a stemming algorithm which is partly from Hazm algorithm used with some modifications. Some suffixes were removed from the algorithm. Then a simple tokenizer is applied to the output so that tokens will be divided. The most important part of the research is the lexicon. I extracted the vocabularies and their synonyms from the book: فرهنگ جامع واژگان مترادف و متضاد زبان فارسي written by: فرجاالله خداپرستي<br>After preprocessing of the texts from the book, and cleaning the data I dumped the words and their synonyms into a python pickle file. Python pickle is a module for serializing and de-serializing python objects. With pickle python objects will be converted into byte streams. Then I wrote a module to load this pickle file and get the input query. It will then process the input query and find the synonyms and show the nearest queries in the sense of meaning.
Cite this paper
Rahmi, A. (2018). Query Expansion. Figshare. https://doi.org/10.6084/m9.figshare.6142769.v1
Rahmi, Adel. “Query Expansion.” Figshare, 2018, https://doi.org/10.6084/m9.figshare.6142769.v1.
Rahmi, Adel. 2018. “Query Expansion.” Figshare. https://doi.org/10.6084/m9.figshare.6142769.v1.
Rahmi, A. (2018) ‘Query Expansion’, Figshare. Available at: https://doi.org/10.6084/m9.figshare.6142769.v1.
@misc{rahmi2018query,
title = {{Query Expansion}},
author = {Rahmi, Adel},
journal = {Figshare},
year = {2018},
doi = {10.6084/m9.figshare.6142769.v1}
}Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.