Comparing LDA and LSA Topic Models for Content-Based Movie Recommendation Systems

Abstract

We propose a plot-based recommendation system, which is based upon an evaluation of similarity between the plot of a video that was watched by a user and a large amount of plots stored in a movie database. Our system is independent from the number of user ratings, thus it is able to propose famous and beloved movies as well as old or unheard movies/programs that are still strongly related to the content of the video the user has watched. The system implements and compares the two Topic Models, Latent Semantic Allocation (LSA) and Latent Dirichlet Allocation (LDA), on a movie database of two hundred thousand plots that has been constructed by integrating different movie databases in a local NoSQL (MongoDB) DBMS. The topic models behaviour has been examined on the basis of standard metrics and user evaluations, performance ssessments with 30 users to compare our tool with a commercial system have been conducted

Business Analytics in Tourism: Uncovering Knowledge from Crowds

“…This can lead to a richer relational structure that reveals latent relations presented between documents and terms (Bergamaschi & Po, 2014). But finding the optimal k is still a challenge.…”

“…But finding the optimal k is still a challenge. Different authors have proposed a number of solutions (Bergamaschi & Po, 2014;Kulkarni, Apte, & Evangelopoulos, 2014;Wild, Stahl, Stermsek, & Neumann, 2005), but many of them refer that this point should be defined empirically for each collection.…”

Topic Detection in Multichannel Italian Newspapers

“…Future work will be focused on comparing the Keygraph algorithm w.r.t. other topic models such as LSA or LDA [4]. Moreover, we would like to investigate how disambiguation techniques might improved the results of Keygraph [3,9].…”

“…The similarity between a community and a document is computed to rank similar documents. The original Keygraph algorithm 4 ) was modified to improve the results of indexing, giving consideration to the hashtags and URLs in news text. The algorithm uses a configuration file that is provided as input and contains numerical parameters useful for clustering (the upcoming words written in italics refer to configuration parameters).…”

Analiza I Obrada Teksta Pomoću Različitih Modela Tema

“…Analiza teksta i modelovanje tema rađeno je u radu [5] koji se bavi poređenjem LDA i LSA modela kako bi se formirao automatski sistem za preporuku filmova. Ovaj rad korišćen je zbog upoređivanja tehnika modelovanja tema i metode evaluacija modela.…”