Smart MM: Smart Movie Management System

Smart MM: Smart Movie Management System

Meenakshi Tripathi (Malaviya National Institute of Technology, India), Saatvik Shah (Malaviya National Institute of Technology, India), Prashant Bahal (Malaviya National Institute of Technology, India), Harsh Sharma (Malaviya National Institute of Technology, India) and Ritika Gupta (Malaviya National Institute of Technology, India)
Copyright: © 2019 |Pages: 27
DOI: 10.4018/978-1-5225-9031-6.ch011

Abstract

Rapid advancements have been made in the field of artificial intelligence in recent years. This has resulted in its adoption in various technologies from medicine to search engines. Existing media management systems have however not yet fully leveraged the power of artificial intelligence (AI) to give users enhanced information apart from basic media metadata. This chapter proposes a smart movie management system which works majorly offline and uses AI to deliver optimum information to the users on four vital tasks. These tasks are multilevel phrase level review polarity, plot and review keywords, a content-based recommendation system, and an emotion recognition system. The complete system works in near-real time with a user-friendly presentation to maximize a user's information gain.
Chapter Preview
Top

Introduction

Media Management systems, both online and offline, have been used ubiquitously in recent years. Online systems provide various utilities such as, integration of relevant information, searching, import, export of metadata, basic personalization, etc. On the other hand, online systems, apart from the above-mentioned properties, also have simplistic recommendation systems along with additional metadata in the form of individual user reviews, links to various critic reviews, crowd sourced rating mechanism etc. The two primary forms of media for which management systems exist are movies and music. Since the focus of the proposed work is towards movies, we will elaborate in terms of the same.

Popular online movie management systems are: MediaCompanion (Media Companion CodePlex, 2016), tinyMediaManager (TinyMediaManager, 2016), Ember Media Manager (EmberMediaManager, 2016) and GCStar (GCStar, 2016). Popularly used online movie management/lookup systems are IMDb, Rotten Tomatoes and MovieCritic etc. Offline management systems suffer from a dearth of features, with the following problems: (1) Only basic metadata is displayed. Such metadata also only includes information about people involved in creating the film (e.g. starcast, directors), genres and ratings. This is too little to effectively critique the movie. (2) Data from multiple sources is not taken into account. (3) No availability of movie reviews or keywords (4) No integrated recommendation systems (5) Movie polarity though, has been considered for positive and negative sentiments of the movies, broader classifications such as happy, sad, anger, fear etc. have not been taken into account. Online management systems provide a few more features but suffer from the following disadvantages: (1) Prerequisite of internet connectivity (2) Non-uniform review ratings i.e. Different people having varying levels of strictness in judgement provide ratings (3) Recommendation system requires users explicit feedback/ratings to generate recommendations (4) Reviews range from short to very long, requiring a large amount of time on part of the reader to get a true picture (5) Manual search required to find relevant movies rather than automatically discovering those which the user already has (6) Plot keywords are determined by non-experts and are thus susceptible to wrong subjectivity.

The proposed Smart Movie Manager System (or SmartMM which is how it is further referred to in the paper) tries to overcome the limitations of the currently available software. It is a huge amelioration from the different movie managing applications available, both online and online. In addition to basic data integration and management, the core of the proposed system is built using AI with the goal of providing maximum amount of relevant information with user-friendly presentation, so as to minimize the time spent on user decision-making.

The core of the proposed system has four primary modules: Firstly, there is an online recommendation engine which suggests movies based on the preferences of the user from a database of over one thousand movies. Secondly, a multilevel (5 levels) and phrase-based review polarity analyzer is built to segment and annotate different sections of the review. Thirdly, a keyword extraction module extracts ten to fifteen relevant keywords from plot summaries and movie reviews. Both, the 2nd and 3rd module help users grasp much more information from metadata in far lesser time and Finally the fourth module is an emotion recognition system, used to extract the percentage of varying emotions in a movie to help in quick categorization of movies depending on their emotional content.

The remaining chapter is divided into four sections. A literature survey of methods related to the four primary modules is covered in next Section. After this, the proposed tool and the techniques used for each module is presented. Evaluation metrics and the performance of the selected techniques under the given use case is discussed in the next Section. Finally, the chapter concludes with some suggestive improvements.

Key Terms in this Chapter

Reviewer: A person who writes critical appraisals of books, plays.

Classifier: Someone who classifies.

Thread: It is an independent set of values for the processor registers.

Semantic: Relating to meaning in language.

Statistical: Relating to the use of statistics.

Metadata: It is a data that describes other data.

Benchmark: A standard against which things may be compared.

Cluster: A group of similar things.

Complete Chapter List

Search this Book:
Reset