Detecting trending products through moving average and sentiment analysis

Trend detection is the process of identifying patterns in data over time. It is used to identify trends in various fields, such as business, finance, marketing, healthcare, and others. For example, in e-commerce, trend detection can be used to identify changes in consumer purchasing behavior, such as an increase in demand for a particular product. Combining trend detection with Sentiment Analysis (SA) can give e-commerce businesses a deeper understanding of customer sentiment over time, allowing them to make more informed decisions about how to improve the customer experience. Our proposed system integrates both the Machine Learning approach and Moving Average (MA) method for sentiment classification of product reviews and trend detection, respectively. To begin, we preprocess the product reviews to facilitate sentiment analysis. We compare the effectiveness of Lexicon-based analysis with machine learning algorithms to identify the optimal approach. The preprocessed reviews are transformed into a vector representation to enable the application of Machine Learning (ML) techniques. Multiple ML algorithms are trained and evaluated based on metrics such as accuracy, F1 score, precision, and recall to determine the best-performing model for sentiment classification. Additionally, we employ clustering techniques to identify the most favorable product. Finally, the MA technique is applied to detect trending products. Our findings indicate that the SVM classifier with BOW achieved an impressive accuracy of 89.26% for sentiment classification. Furthermore, by utilizing K-Means clustering, we obtained a high silhouette score, outperforming other clustering methods, in identifying trending products. Our study highlights that the combination of sentiment analysis and trend detection can provide valuable insights for e-commerce businesses.