Data Architectures for Information Retrieval and Web Intelligence
Mål og innhald
The search and content optimisation problems are important research and development themes in various Web-based applications. Crucial challenges are to find the best way to optimize keywords and to develop personally relevant content from the widely diverse Web. The question is if users could be steered to content of high relevance for them. Web analytics and intelligence provide methods and tools that could help accomplishing such user expectations.
The course focuses on the information retrieval as goal and as a part of the process of extracting knowledge from the Web resources. It goes beyond the click-stream analysis and explores techniques for discovery of facts and new patterns. The course also provides insights into Web dynamic looking at challenges of constant change and how it affects Web content, size, topology and use.
Methods presented in the course could be used for small as well as fairly comprehensive business intelligence projects. Web based technologies are not only promoting a special brand of data culture within a company; they impact the way business is being promoted and conducted.
Specific study goals are to:
- Learn about the performance of data collection methodologies
- Explore the history, theory, and practice of faceted search
- Discover how to identify valuable search metrics
- Explore text processing techniques including text mining
- Optimize the organizational structure and choose the right analytics tool
- Understand and apply advanced analytics concepts, including SEM/PPC analysis, the power of segmentation, conversion-rate best practices, and others
- Learn the key ingredients of a great experimentation and testing platform
- Use competitive-intelligence analysis to glean insights and drive actions
- Explore quick-start solutions for blogs and e-commerce, support, and small business websites
- Learn to use web analytics data to help make strategic decisions and set business goals.
Upon completion of the course the candidate should be able to
- demonstrate an understanding for concepts and theories in the field of Information Retrieval, and explain and discuss how information retrieval concepts and methods can be applied for implementing evidence based decisions and recommender systems.
- use methods based on mathematics and artificial intelligence to retrieve information from a variety of web-based sources, databases, and social media.
- use techniques and tools to do research in the area of information retrieval and text mining, and to apply methods of Information Retrieval and understand their feasibility and performance.
- develop a real-world application and explain the choice of technologies to solve particular problems.
- identify relevant research and trends in Information Retrieval for improving search accuracy from user feedback, recommendation systems, scientific literature, as well as new research problems.
- work in a team to produce a functioning system involving Information Retrieval.
- produce creative solutions based on current research and industry trends in Search Technology.
Krav til forkunnskapar
Bachelor´s degree in Information Science or equivalent.
Krav til studierett
Master in information science. Students admitted to other master´s programs and international exchange students may also be qualified to apply for the course.
Undervisningsformer og omfang av organisert undervisning
Lectures (approximately 40 hours), work on individual and group assignments, presentations, and discussions. Normally 5 hours once every second week until mid or late May.
There will be mandatory assignments which must be completed and approved. The approved mandatory assignments cannot be carried over to other semesters.
Mandatory participation: Attendance at 80 % of course sessions is mandatory.
Essay on selected topics
The grading system has a descending scale from A to E for passes and F for fail.
INFO323 is evaluated by students every three years, by the Department every year.