Abstract
Numerous individuals use GSMArena to find the decent mobiles. In any case, with just a general rating for every mobiles, GSMArena offers insufficient data for freely making a decision about its different angles, for example, condition, company or model. Regardless, with just a general rating for every mobile, GSMArena offers insufficient data for autonomously making a decision about its different viewpoints. In this project, the goal is to point out demand of customers from a large amount of reviews, with high dimensionality. These themes can give significant bits of knowledge to reviews about what clients care about so as to expand their GSMArena appraisals, which straightforwardly influences their income. In proposed, would like to predict ratings of mobiles on GSMArena and popularity change based on mobile features, such as available mobiles, price level, etc. It cannot only shed lights on what customers value the most about a mobile, but also provide suggestions on what feature combinations one should choose when opening a new mobile, and how likely this mobile can succeed. Uses several machine learning methods including logistic regression to make relevant predictions. Preprocessing, Data Extraction, Data Analytics, Visualization, Result, Evaluation, Decision making. While strategic relapse performs superior to the others, forecasts from every one of the strategies are a long way from immaculate. This suggests the potential improvement of more information and increasingly fit procedures. It enables individuals to get a comprehensive view on a specific mobiles dependent on its essential data, pictures, surveys, etc. The rating of reviews on GSMArena likewise turns into a critical marker of whether a mobiles is effective and well known.