Abstract

We suggest to use the linked users through social networking sites and e-commerce websites as a link to map users’ social networking structures to added feature demonstration for product recommendation. In detailed, we suggest wisdom both users’ and products’ feature illustrations (called user embeddings and product embeddings, individually) from data collected from e-commerce websites using repeated neural networks and then apply a improved gradient boosting trees technique to change users’ social networking structures into user embeddings. We then improve a feature-based matrix factorization method which can leverage the learnt user embeddings for cold-start product recommendation.

Authors and Affiliations

Kollimarla Sravanthi| Dept. of CSE,Vellaga Nageswara Rao College Of Engineering,Ponnur, Andhra Pradesh 522124, K Sheeba Rani| Dept. of CSE,Vellaga Nageswara Rao College Of Engineering,Ponnur, Andhra Pradesh 522124

Keywords

Related Articles

Under water Image preprocessing by Average filter and a comparison study

Underwater image pre-processing is extremely essential because, under water images suffer from quality degradation due to low transmission of light. When an underwater image is captured, pre-processing is necessary t...

IoT Based Smart Trash /Garbage Collection System

Waste Management is one of the major problem in our day to day life. The current observation is that trash-bin is not efficiently managed, it is not emptied from time to time thus leading to a critical situation whic...

Simulation and Application of Maximum Power Tracking Scheme in a Novel Excitation Synchronous Wind Power Generator (ESWPG) by using Fuzzy Logic Controller

This paper proposes a novel excitation synchronous wind power generator (ESWPG) with a maximum power tracking scheme by using Fuzzy Logic Controller (FLC). The excitation synchronous generator and servo motor rotor s...

Stashing Scope Afore Encoding For Revocable Data Spanking In Encoded Images

Nowadays, information is rapidly available through the Internet. Companies have the ability to communicate with a worldwide audience through the World Wide Web. The extraction is performed in the reverse order as the...

Web-based social networking foundations makes clients unable to fittingly control to whom these items are really shared or not. Computational components that can combine the protection inclinations of numerous clients...

Download PDF file
  • EP ID EP17000
  • DOI -
  • Views 230
  • Downloads 3

How To Cite

Kollimarla Sravanthi, K Sheeba Rani (2017). . International Journal of Science Engineering and Advance Technology, 5(7), 827-829. https://europub.co.uk/articles/-A-17000