• Perugu Swathi
• Mr. G.Prudvi Raj
• Aseen Babu
• M.Haritha

Today's commerce is heavily influenced by the power of online reviews, with consumers relying heavily on them to make purchasing decisions. However, not all reviews can be trusted as some individuals or organizations may manipulate them for their own benefit. In this study, we explore the efficacy of two text mining techniques on a collection of hotel review data and present innovative semi-supervised and supervised text mining models to detect misleading online reviews. In the realm of purchasing and business, online assessments have become imperative in determining individuals' choices. Regrettably, the prevalence of fraudulent appraisals, either manually crafted or artificially produced, has become an issue. This project endeavors to rectify this predicament by constructing a categorizer that forecasts the validity of an evaluation through analyzing both the review text and the conduct of the user. The categorizer was prepared using various learning methodologies, including Logistic Regression, Multinomial Naive Bayes, K-Nearest Neighbor (KNN), Random Forest Classifier, Convolutional Neural Network (CNN), and CNN with Long-Short Term Memory (LSTM). The results indicated that Logistic Regression and KNN demonstrated the most effectiveness, yielding approximately 60-64% accuracy. Conversely, Naive Bayes and Random Forest Classifier showed a lower accuracy of 50%, while CNN-LSTM had a lower accuracy of 21%, but a noteworthy recall of 0.82.

Using Semi-Supervised and Supervised Learning for Fake Online Review Detection

"Using Semi-Supervised and Supervised Learning for Fake Online Review Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 3, page no.e471-e477, March-2023, Available :http://www.jetir.org/papers/JETIR2303458.pdf

"Using Semi-Supervised and Supervised Learning for Fake Online Review Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 3, page no. ppe471-e477, March-2023, Available at : http://www.jetir.org/papers/JETIR2303458.pdf