Loading...
Persian Query Corrector Based on Deep Learning Networks (with Emphasis on Spatial Queries)
Shahrivari, Vahid | 2021
258
Viewed
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 54192 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s): Izadi, Mohammad
- Abstract:
- The main spectrum of a location search engine is retrieving the locations that are most relevant to the user query. Geographic and spatial queries usually consist of a series of keywords that express the user's needs. The geographic search engine should retrieve the locations associated with the user request and then rank the retrieved results according to their relevance to the user request. In some cases, user requests may contain spelling errors that can greatly affect the results retrieved. Spell correction is automatic in the back spectrum recovery system. Spelling correction is the task of automatically recovering the intended text from a misspelled text. Therefore, spelling correction systems are a very important tool in natural text processing because they significantly improve the performance of language processing systems and Information Retrieval systems. that improve their performance on misspelled text. Recently, deep learning-based Natural Language Processing (NLP) models have generated promising results on natural language processing. Designing an automated spelling correction system requires knowledge of the factors that cause a spelling error. Numerous traditional models for spelling correction have been proposed, however, more recently, natural language processing models based on deep learning have proposed stunning results in natural language processing. In this dissertation, we present a spelling correction model based on deep learning techniques. The evaluation results indicate an accuracy of 86.3\% and a recall of 54.8\% on the artificial data set and accuracy of 83\% on user queries. It works better than some existing methods. We also provide a user's query model and ranking model for enhancing the performance of geographic searches. The dataset used in this dissertation is from the OpenStreetMap database. While the experiments of this research have been done in the field of the Persian language, the methods presented in this dissertation can be easily applied to any other language
- Keywords:
- Natural Language Processing ; Deep Learning ; Spell Correction ; Query Preprocessing ; Query Spell Correction