Certain investigation on document Sentiment analysis and analyzing Discussions on online forum posts Using intelligent data mining Techniques

Abstract

Sentiment Analysis or Opinion Mining is a Natural Language Processing method used to assess whether data is positive, negative or neutral. Sentiment Analysis is also conducted on textual information to help companies track brand and product sentiment in consumer reviews, and recognize customer demands. In recent days, Web search engine tools have been supporting people in the journey of information. From the record and ordering frameworks to the cutting edge million-results-under-a-second nature of inquiry frameworks are truly long voyages. A large portion of the web indexes accepts crude watchwords as a trigger to create the information from the huge stores for the most part. Several users provide reviews but only a few parameters are proven to be significant. The recommendation or opinion given to the potential user is based on the type of feedback. newlineFor the purpose of the experiment, many reviews were gathered and sentiment analysis was performed. The opinion polarity has been calculated using the automatic weighted method. The suggested solution outperforms the current methodology by applying the automatic weight. Using the proposed Multinomial Naive Bayes SVM (MNBSVM) classifier method the user reviews of mobile phones has been predicted with the help of positive and negative count reviews. The previous techniques focus on the sentiment scores generated by the class of documents upon collection, and it is a time consuming process. In order to overcome these problems, the proposed mechanism of Collective Parallel Cluster Algorithm has been developed. The proposed model performs well in terms of accuracy, lesser human intervention with efficiency and feasibility analysis. newline