Implicit Preferences Discovery for Biography Recommender System Using Twitter

Graduate students always face a dilemma when it comes to choosing electives every semester. Different data sets have been used in order to avoid this confusion and chaos. In order to help them choose their subjects as per their capability, we use data mining and natural language processing techniques that helps in conversion of human-readable format to machine-readable format, both of which are vastly emerging fields to propose a collaborative recommendation system.

Recommendation System for Students’ Course Selection

Recommender systems, which suggest relevant products to internet users, have become an integral part of our daily lives. The factors responsible for their success from the different stakeholder perspectives, however, have never been thoroughly investigated. This study proposes a novel model for measuring the success of recommender systems that consolidates different success factors. The model is a modified version of the DeLone and McLean Information Systems Success Model with trust as an additional latent variable. The model was evaluated in an empirical study with PLS-SEM. The proposed model exhibits a high predictive power and all structural paths were significant. The integration of trust is an important contribution as the path between information quality and trust yielded the highest path coefficient. The proposed model can be used by recommendation system providers to explain and predict the successful use of the systems and to improve business processes.

https://ieeexplore.ieee.org/ielx7/6287639/9668973/09734033.pdf

Measuring the Success of Recommender Systems: A PLS-SEM Approach

Social media can be perceived as being a predominantly visual platform (as we will explore in ► Chaps. 15 and 16), due to the high levels of engagement that images and video can generate across channels (Brubaker & Wilson, 2018; Marshall, 2018). Yet, underestimating the power and necessity of well-written copy can be an error made to the detriment of a social media strategy. Generally, audiences consume text online in snack-sized portions reading an average of 28 words per website visit (Weinreich, Obendorf, Herder, & Mayer, 2008). Therefore, the written word can play a considerable role in cutting through the wide range of content a target audience scrolls past on a daily basis in social media’s attention economy (Quesenberry, 2018).

Writing for Social Media

Recommender systems have become essential in several domains to deal with the problem of information overload. Collaborative filtering is one of the most popularly used paradigm of recommender systems for over a decade. The personalized recommender systems use past preference history of the users to make future recommendations for them. The cold start problem of recommender system concerns with the personalized recommendation to the users having no or few past history. In this work we propose an approach to learn implicit user preferences by making use of YouTube Video Tags. The profile of a new user is created from his/her preferences in watching the YouTube videos. This profile is generic and may be used for a wide variety of domains of recommender systems. In this work we have used it for a biography recommender system. However this may be used for several other types of recommender system.

Rahul Upadhyay

Papers

Learning User Preferences for Recommender System Using YouTube Videos Tags