Multimedia Information Retrieval

With the proliferation of text and multimedia information, users are now able to find answers to almost any questions on the Web. However, they are often bewildered by the huge amount of information available to them. Information retrieval (IR) involves the processing and analysis of such unstructured information in order to return the most relevant set of information in answer to the user’s queries.

Primaries

Students interested in learning in-depth knowledge about this focus area would start with CS2108 Introduction to Media Computing to learn about the basics of multimedia data representation and processing techniques as well as CS3245 Information Retrieval to learn about the key data structures and algorithms for document indexing and retrieval. Words based indexing and retrieval, particularly for unstructured text, typically yields unsatisfying search results, and thus it is important to adopt natural language processing techniques to understand the content of a given document. CS4248 Natural Language Processing serves as a focused introduction to this topic. CS4242 Social Media Computing looks at information retrieval in the context of social media, where social media posts such as tweets and comments play an increasingly important role in understanding the users. Rounding up the list of primaries, CS4347 Sound and Music Computing delves into understanding, retrieving, recommending content in the context of sound and music, a topic that plays a crucial role in the success of online music streaming platforms.

Electives

Two advanced electives in this focus areas are CS5241 Speech Processing, which complements CS4347 by focusing on human speech understanding, and CS5246 Text Mining, which complements CS4242 and CS4248 on the topics of text classification, text clustering, sentiment analysis, text summarization, information extraction, and question answering.