1 / 13

Funzic have fun…!

Funzic have fun…!. Harish Kambam Pavankanth Mukthi Chandrasekhar Manda. Agenda. Demo Introduction Business Case Functionalities Architecture Data Mining Novelty Team Contribution. Introduction. Necessity for Social networks to be more specific

xena
Download Presentation

Funzic have fun…!

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Funzichave fun…! Harish Kambam Pavankanth Mukthi Chandrasekhar Manda

  2. Agenda • Demo • Introduction • Business Case • Functionalities • Architecture • Data Mining • Novelty • Team Contribution

  3. Introduction • Necessity for Social networks to be more specific • Identifying friends, friends within the friends’ networks based on the Similarities (music) and likes as per Social Networking profile

  4. Business Case • To provide a recommendation system that • Suggests Music and Music Artists based on user’s preferences & Social Networking presence • Revenue: Facebook App, Android App, Google Adwords

  5. Functionalities • Search Music Artists • Search Movies • Artist Recommendation • User Input • Facebook Profile Information

  6. Architecture

  7. Technical Details • Latest Facebook Connect Technology • Over 200,000 artist tags fetched from last.fm and spiders • Music Dataset (fetched via API calls from Last.fm) • Top 10000 artists, 87000 tracks, 45000 albums • Movie Dataset (fetched using text spiders from Imdb.com/UCB.edu) • Movie titles, ratings, soundtracks, book adaptations • Movie data as recent as April 2010 • Mashup of Music/Movie information

  8. Competitor Analysis • TweetURMusic • Twitter based music sharing portal • No data mining component • No Videos • MusicMazaa • Mashups – Photos, Videos, Sports, Blogs • No Dataming and Recommendation System • Funzic! • 9 Top APIs – all rated best in their own regard • Data Mining Component and Recommendation System • Tons of Data

  9. Data Mining We have We did • Over 10,000 Popular Artists (Clean data) – (over 5,000 unique tags) • All Movie Soundtracks* for Artists • All Book adaptations* for Artists • Tag dictionary • Eliminated redundancy • Normalization of Data • Jaccard function • Similarity for attributes with user query • Recommendation System • Facebook Profile • User Input *upto April 2010

  10. Novelty • One click recommendation for Similar friends, Artist Suggestions based on Facebook profile • One stop information site for Music, Music related to Movies (Based on listed Sound tracks) • Fully functional Web 2.0 enabled Mashup site with YouTube, Flickr, RSS feeds, Wikipedia, Play.me, Amazon, last.fm, IMDB, Twitter & Facebook • User friendly, Interactive and fully functional website • Target user base – Social Networking platform of Facebook

  11. Team Contribution • Pavankanth • Facebook API, Play.me API, YouTube API, Flickr API, UCB Dataset spidering • Web application development & Architecture • Harish • Web design, User Interface development, CSS, JavaScript & JQuery Effects • Wikipedia API, RSS Feeds • Chandrasekhar • Last.fm Crawler, Parser, IMDB Parser, Amazon API • Data base design, Data mining

  12. Data Mining (Additional) • Normalization • Useful to normalize errors when population parameters are known

  13. Data Mining (Additional) • Jaccard Function (Similarity Scores) • Each user input, Each Artist, Each Friend on Facebook is weighed for 15 different attributes • Jaccard Function is utilized to identify • Similarity between the User inputs & Artists of Funz!c dataset

More Related