TAG-BASED RECOMMENDATION SYSTEMS

Abstract

With the growth in the use of the internet the information available online is huge, browsing through this information manually is a tedious task. To simplify this Recommendation Systems were introduced that served us with the relevant information easily. Recommendation Systems use the user's profile that has data related to his past interactions, items description, and preferences of similar user to fetch information. Social tagging is also an advancement in the web. It allows user to describe the item using keywords known as tags. These tags capture the semantic value of the item better than any other form of rating and hence are able to assist recommend with a higher accuracy. Tag information can be used to find similar users .Collaborative filtering helps in finding items that match our preferences and also users that have a similar interest as us. This information can be used to generate item recommendation for a particular user. We will look in to these systems and their properties in greater detail in this thesis report.