Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

RECOMMENDATION SYSTEMS IN SOCIAL NETWORK

No description
by

thalia lajara pomares

on 5 January 2018

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of RECOMMENDATION SYSTEMS IN SOCIAL NETWORK

INTRODUCTION
Challenges
Literature review
MSc Big Data Technologies
IT Professional Issues and
Project Methods

Thalia Lajara
Research Report Presentation
Smart classification of Social Network news feed
How achieve our goals:

A classifier will be developed. Classification of fresh data into categories based on trained data
Use of recommendation systems to predict the "rating" or "preference" that a user would assign to each article

Initial proposal for news classification
LAST NEWS
Entertainment
Politics
Food
Sports
Social - My friends news
Find the right techniques to analyze, classify and recommend news.

States of art
GMail filter - classification
EXPERIMENTAL EVALUATION WITH A SIMPLE CASE STUDY
Facebook
Technical evaluation
Study of the current Facebook news structure
Methodologies used versus desired methodologies
Search for the appropriate classification algorithm
Study of the best recommendation approach:
Collaborative filtering, Content-based filtering, Hybrid systems
EXPERIMENTAL EVALUATION WITH A SIMPLE CASE STUDY
Facebook
PROTOTYPE AND EXPERIMENT
Use of R
Recommendation
Smart classification of Social Network news feed
INTRODUCTION
Different languages
Data collection and extraction
Provide a better structure for Facebook :
(Non-categorized news to categorized news)
Find the right tools to manage Big Data from different sources.
Recommendation systems approaches
Support Vector Machine (SVM)
Build a classifier
EXPERIMENTAL EVALUATION WITH A SIMPLE CASE STUDY
Facebook
PROTOTYPE AND EXPERIMENT
Sampling:
Data collection
Data extraction
Classification into known categories
Filtering
Health & Fitness
Worldwide
Categories
END
Smart classification of Social Network news feed
Classification
To which category (sub-population) a new observation belongs

Training set of data is needed

The different categories are known
Use of computing tools
Rationale for developing a news classifier
Lack of power of choice of the subject of interest in each moment - Not human interaction
The news are not classified by topic in social networks

Optimization of the time we spend on social networks: find more relevant data in less time
Increase the amount of relevant news according to our preferences. More quality time in the platform - Happier users
Research aims:

Manage different types of data
Recommendation + social network
Filtering + Social Network
SMART CLASSIFICATION IN SOCIAL NETWORK NEWS FEED
News classification
ALGORITHMS AND METHODS
ALGORITHMS AND METHODS
Full transcript