Product Recommendation Using Microblogging Information Social Media

0
583
Connecting Social Media to E-Commerce: Cold-Start Product Recommendation Using Microblogging Information

Product Recommendation Using Microblogging Information Social Media

Abstract:

Product Recommendation Using Microblogging Information Social Media Decade, the boundaries between e-commerce and social networking have become increasingly blurred. Lots of e-commerce web Application support the process of social login where users can sign on the websites using their social network username and password authentication such as their Twitter or Facebook accounts. Social Network users can also post their newly purchased products on microblogs with links to the e-commerce product web pages. In this paper, we propose a novel solution for cross-site cold-start product recommendation. We aim to recommend ecommerce product from e-commerce websites to users at social networking websites in “cold-start” situations. Coldstart situation is a problem which has rarely been explored before.

A major challenge is how to leverage knowledge extracted from social networking sites for cross-site cold-start product recommendation. We propose to use the linked users across social networking sites and e-commerce websites as a bridge to map users’ social networking features to another feature representation for product recommendation. In specific, we propose learning both users’ and products’ feature representations from data collected from e-commerce websites using recurrent neural networks and then apply a modified gradient boosting trees method to transform users’ social networking features into user embeddings. We then develop a feature-based matrix factorization approach which can leverage the learnt user embeddings for cold-start product recommendation. Experimental results on a large dataset constructed from the microblogging service FACEBOOK and the largest e-commerce website AMAZON have shown the effectiveness of our proposed framework.

INTRODUCTION

In recent years, the boundaries between e-commerce and social networking have become increasingly blurred. Ecommerce websites such as eBay features many of the characteristics of social networks, including real-time status updates and interactions between its buyers and sellers. Some e-commerce websites also support the mechanism of social login, which allows new users to sign in with their existing login information from social networking services such as Facebook, Twitter or Google+. To address this challenge, we propose to utilize the linked users across convivial networking sites and e-commerce websites (users who have gregarious networking accounts and have made purchases on e-commerce websites) as a bridge to map users’ gregarious networking features to latent features for product recommendation.

In concrete, we propose learning both users’ and products’ feature representations (called utilizer embeddings and product embeddings, respectively) from data amassed from ecommerce websites utilizing recurrent neural networks and then apply a modified gradient boosting trees method to transform users’ gregarious networking features into utilizer embeddings. We then develop a feature predicated matrix factorization approach which can leverage the learnt utilizer embeddings for cold-start product recommendation. We built our dataset from the most immensely colossal Chinese micro blogging accommodation SINA WEIBO2 and the most astronomically immense Chinese B2C e-commerce website, containing a total of 20,638 linked users. The experimental results on the dataset have shown the feasibility and the efficacy of our proposed framework. Our major contributions are summarized below:  

  • We formulate a novel quandary of recommending products from an e-commerce website to convivial networking users in “cold-start” situations. To the best of our erudition, it has been infrequently studied afore.
  • We propose to apply the recurrent neural networks for learning correlated feature representations for both users and products from data amassed from an e-commerce website.
  • We propose a modified gradient boosting trees method to transform users’ micro blogging attributes to latent feature representation which can be facilely incorporated for product recommendation.
  • We propose and instantiate a feature-predicated matrix factorization approach by incorporating utilizer and product features for cold-start product recommendation.

Related Work:

In our recommendation system for recommending colleges, we decided to take a different approach to the problem. Existing approaches tend to focus on user-item matrix techniques and neighbourhood approach, and their models reflect this line of thinking. We still do similarity calculations, but in a different way for recommending colleges as venues. There are some concepts that we use, which are common to most currently existing recommendation colleges. our project systems rely on information derived from the online of users, such as opinions or ratings, to form predictions, or produce recommendation of colleges . Existing collaborative filtering techniques involve generating a user item in fake matrix, from which recommendation results could be derived.

System Configuration:

H/W System Configuration:-

Processor          : Pentium IV

Speed               : 1 Ghz

RAM                  : 512 MB (min)

Hard Disk          : 20GB

Keyboard           : Standard Keyboard

Mouse               : Two or Three Button Mouse

Monitor             : LCD/LED Monitor

S/W System Configuration:-

Operating System               : Windows XP/7

Programming Language       : Java/J2EE

Software Version                 : JDK 1.7 or above

Database                            : MYSQL

 

Module Description:

Twitter Login

Twitter has evolved from a micro blogging service into a popular social messaging platform, it has been instrumental in providing the “pulse” on news and events across the globe. In addition to its widespread acceptance among the news media and entertainment industry, Twitter has also become a popular social media marketing tool and a great way to communicate with both friends and co-workers. It is many different things to many different people. It can be used by a family to keep in touch, or a company to coordinate business, or the media to keep people informed or a writer to build up a fan base. Twitter is micro-blogging. It is social messaging. It is an event coordinator, a business tool, a news reporting service and a marketing utility.

Register: When you join Twitter from your phone, we collect all of your updates on the web. When you’re ready to activate your web account, all you have to do is add an email address and a password to complete your profile and log in. Enter your full details of the signup page.  Enter your phone number when prompted.  Twitter will send a verification code to your phone.

  • When you get that text message, enter the code. You’ll then be asked to create an account by entering your email address, a password, and a name for your account.
  • Your username will already be entered, since you created this when you signed up via SMS.
  • Click Create my account as shown below, and you’re all done! Twitter will walk you through finding some friends you may know on Twitter, then will direct you to your home page 2.

Followers: A number of companies have sprung up offering to help people and companies grow their Twitter following the easy way by forking over a little cash. The usefulness of a large Twitter following isn’t simply a number. Having 10,000 followers won’t do you much good if only half of them are real people and most of them aren’t interested in anything you tweet. It may change the number of followers at the top of the screen, but it will have little effect on how many people actually follow what you say. The entire point of growing a large Twitter following is to communicate with people who are interested in your topic, your company or your brand. They want to read what you write, retweet what you tweet and check out the links you post. This is what makes a Twitter following valuable. And a Twitter following that simply ignores your tweets simply isn’t worth it.

Follow: Twitter can be a great way to promote you, your brand or your company. Actors, writers, sports stars and others are turning to Twitter as a way to connect with fans and promote themselves to millions of people across the globe. Twitter is also a great way for bloggers to gain more traction in the blogosphere and get noticed. But how do you build up a Twitter following? These simple tips will help you get started.

EXISTING SYSTEM:

  • Most studies only focus on constructing solutions within certain e-commerce websites and mainly utilise users’ historical transaction records. To the best of our knowledge, cross-site cold-start product recommendation has been rarely studied before.
  • There has also been a large body of research work focusing specifically on the cold-start recommendation problem.
  • Seroussi et al. proposed to make use of the information from users’ public profiles and topics extracted from user generated content into a matrix factorization model for new users’ rating prediction.
  • Zhang et al. propose a semi-supervised ensemble learning algorithm.
  • Schein proposed a method by combining content and collaborative data under a single probabilistic framework.
  • Lin et al. addressed the cold-start problem for App recommendation by using the social information.

DISADVANTAGES OF EXISTING SYSTEM:

  • They only focus on brand or category-level purchase preference based on a trained classifier, which cannot be directly applied to our cross-site cold start product recommendation task.
  • Their features only include gender, age and Facebook likes, as opposed to a wide range of features explored in our approach.
  • They do not consider how to transfer heterogeneous information from social media websites into a form that is ready for use on the e-commerce side, which is the key to address the cross-site cold-start recommendation problem.

PROPOSED SYSTEM:

  • In this paper, we study an interesting problem of recommending products from e-commerce websites to users at social networking sites who do not have historical purchase records, i.e., in “cold-start” situations. We called this problem cross-site cold-start product recommendation.
  • In our problem setting here, only the users’ social networking information is available and it is a challenging task to transform the social networking information into latent user features which can be effectively used for product recommendation. To address this challenge, we propose to use the linked users across social networking sites and e-commerce websites (users who have social networking accounts and have made purchases on e-commerce websites) as a bridge to map users’ social networking features to latent features for product recommendation.
  • In specific, we propose learning both users’ and products’ feature representations (called user embeddings and product embeddings, respectively) from data collected from e-commerce websites using recurrent neural networks and then apply a modified gradient boosting trees method to transform users’ social networking features into user embeddings.
  • We then develop a feature-based matrix factorization approach which can leverage the learnt user embeddings for cold start product recommendation.

ADVANTAGES OF PROPOSED SYSTEM:

  • Our proposed framework is indeed effective in addressing the cross-site cold-start product recommendation problem.
  • We believe that our study will have profound impact on both research and industry communities.
  • We formulate a novel problem of recommending products from an e-commerce website to social networking users in “cold-start” situations.
  • To the best of our knowledge, it has been rarely studied before.
  • We propose to apply the recurrent neural networks for learning correlated feature representations for both users and products from data collected from an e-commerce website.
  • We propose a modified gradient boosting trees method to transform users’ microblogging attributes to latent feature representation which can be easily incorporated for product recommendation.
  • We propose and instantiate a feature-based matrix factorization approach by incorporating user and product features for cold-start product recommendation

ADVANTAGES

  1. The proposed method does not rely on the textual contents of social network posts. It is robust to rephrasing and it can be applied to the case where topics are concerned the information other than texts, such as images, videos, audio and so on.
  2. The proposed link anomaly based method performed even better than the keyword based methods on “NASA” and “BBC” dataset.

SCOPE OF PROJECT

  1. Easy to advertise product exploitation social networking web site.
  2. Increase the interaction between user and social networking website.
  3. We believe that our study can have profound impact on each analysis and business communities.
  4. We propose a changed gradient boosting trees technique to rework users’ microblogging attributes to latent feature illustration which may be simply incorporated for product recommendation.
  5. We tend to propose and instantiate a feature-based matrix resolving approach by incorporating user and merchandise options for cold-start product recommendation.

Conclusion and further work

In Product Recommendation Using Microblogging Information Social Media paper, we have concentrated on a novel issue, cross-site cool begin item suggestion, i.e., prescribing items from e-trade sites to micro-blogging clients without authentic buy records. Our primary thought is that on the e-trade sites, clients and items can be spoken to in the same dormant element space through element learning with the repetitive neural systems.

Utilizing an arrangement of connected clients crosswise over both e-trade sites and long range interpersonal communication destinations as an extension, we can learn include mapping capacities utilizing a changed angle boosting trees technique, which maps clients’ qualities extricated from long range informal communication locales onto highlight representations gained from e-business sites. The mapped client components can be adequately joined into a include based network factorization approach for cold start item proposal. We have built a vast dataset from WEIBO and JINGDONG. The outcomes demonstrate that our proposed system is without a doubt compelling in tending to the cross-site icy begin item suggestion issue. We trust that our study will have significant effect on both research and industry groups.