A comparison of Bangladeshi daily news portals’ various clustering techniques for online revenue generation

The newspaper is often regarded as one of the most effective means of acquiring news and remaining informed about both local and global occurrences. Technological improvements have contributed to the comparable popularity of online newspapers and traditional magazines. The datasets used in this analysis were obtained from the online newspaper sections of two well-known and respected English and Bengali daily newspapers in Bangladesh, “The Daily Star” and “Prothom Alo”. This study examines several aspects of the analysis, organizing them into distinct categories. Similarly, online newspapers are using Google Analytics to monitor both overall views and unique page views. Consequently, this industry has been the focal point of various scholarly publications. In this study, different clustering techniques were used, and it was observed that in both newspapers, Spectral Clustering performed better than the other clustering techniques. Spectral Clustering obtained $\mathbf{8 9 \%}$ and $\mathbf{8 3 \%}$ Silhouette Scores for the Daily Star and Prothom Alo respectively. Achieving this goal will allow newspapers to obtain significant insights on market trends and customer behavior, as well as increase revenues and attract more readers.