Developing a framework for analysing web data to generate recommendation

From the advent of the Internet, data contained in the web is increasing exponentially and the number of e-commerce companies is also increasing significantly. Nowadays, we can find that many e-commerce companies selling same products and / or services at different prices. For a customer it is difficult to find an e-commerce company which will be best suited for him. Moreover, it is time consuming and tedious to search for a product in different online sites. For reducing this difficulty, in this paper, we develop a system that can extract web data from different e-commerce sites even though the language of the websites are different. We then analyse the extracted data and recommend best products / services from these sites to the users. For the experimental evaluation of our system, we consider two different languages: English and Bangla and extract books data from different online book stores considering these two languages and recommend books to the users. From the experimental results, we can find that our system can help the users in selection of their products easily and efficiently.