Computer Science > Machine Learning
[Submitted on 7 Jan 2016 (this version), latest version 15 Jun 2016 (v3)]
Title:From Word Embeddings to Item Recommendation
View PDFAbstract:Social network platforms can archive data produced by their users and re-use the data to serve the users better. One of the services that these platforms provide is the recommendation service. Recommendation systems can predict the future preferences of the users using various different techniques. One of the most popular technique for recommendation is matrix-factorization, which uses low-rank approximation of input data. Similarly word embedding methods from natural language processing literature learn low-dimensional vector space representation of input elements. Noticing the similarities among word embedding and matrix factorization techniques and based on the previous works that apply techniques from text processing for recommendation, Word2Vec's skip-gram technique is employed to make recommendations. Unlike previous works that use Word2Vec for recommendation, non-textual features are used. The aim of this work is to make recommendation on next check-in venues and a Foursquare check-in dataset is used for this purpose. The results showed that use of vector space representations of items modelled by skip-gram technique is promising for making recommendations.
Submission history
From: Makbule Gulcin Ozsoy [view email][v1] Thu, 7 Jan 2016 00:09:37 UTC (890 KB)
[v2] Sun, 6 Mar 2016 16:09:10 UTC (890 KB)
[v3] Wed, 15 Jun 2016 08:07:36 UTC (922 KB)
Current browse context:
cs.LG
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
IArxiv Recommender
(What is IArxiv?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.