Aspect Based Opinion Mining & Sentiment Analysis

Autor: Alka Londhe, P. V. R. D. Prasada Rao
Rok vydání: 2022
Předmět:
Zdroj: International Journal on Recent and Innovation Trends in Computing and Communication. 10:131-142
ISSN: 2321-8169
Popis: Opinion mining is a relatively new field that refers to the practice of collecting feedback in the form of online reviews and ratings left by users on various topics. Researchers are now able to monitor the states of consciousness of individuals in real-time because to this development. Just lately, a number of research papers for sentiment analysis were implemented, each of which was based on a unique categorization and ranking procedure. However, the amount of time necessary for the newline performing class has not decreased in any way. Sentiment Sensitivity newline word list SST was provided as a solution to the problem of function mismatch in the go-domain sentiment class across the source area and the target domain; however, achieving improved accuracy and identifying distributional similarities of words became less effective as time went on. Hidden Markov’s persistent development may be seen at the beginning. Cosine In order to achieve more effective and clean pre-processing, a method that is conceptually quite similar to HM-CPCS has been devised. The HM-CPCS methodology, which has recently been suggested, makes use of the POS tagger, a variant of which is based on the Hidden Markov algorithm. Evaluations are created using data from a wide variety of different domains. Similar to a newline, the tags that come before and after it compute the possibility of transitions and the existence of the term newline among the tags in order to increase capability. This is done in order to improve capability.
Databáze: OpenAIRE