CS 430 / INFO 430 Information Retrieval

1. CS 430 / INFO 430 Information Retrieval Lecture 23 Non-Textual Materials 2

2. Course Administration Assignment 3 Grades and comments will be sent out tomorrow Assignment 4 has been posted

3. Automatic Creation of Surrogates for Non-textual Materials Discovery of non-textual materials usually requires surrogates • How far can these surrogates be created automatically? • Automatically created surrogates are much less expensive than manually created, but have high error rates. • If surrogates have high rates of error, is it possible to have effective information discovery?

4. Example: Informedia Digital Video Library Collections: Segments of video programs, e.g., TV and radio news and documentary broadcasts. Cable Network News, British Open University, WQED television. Segmentation: Automatically broken into short segments of video, such as the individual items in a news broadcast. Size: More than 4,000 hours, 2 terabyte. Objective:Research into automatic methods for organizing and retrieving information from video. Funding: NSF, DARPA, NASA and others. Principal investigator: Howard Wactlar (Carnegie Mellon University).

5. Informedia Digital Video Library History • Carnegie Mellon has broad research programs in speech recognition, image recognition, natural language processing. • 1994. Basic mock-up demonstrated the general concept of a system using speech recognition to build an index from a sound track matched against spoken queries. (DARPA funded.) • 1994-1998. Informedia developed the concept of multi-modal information discovery with a series of users interface experiments. (NSF/DARPA/NASA Digital Libraries Initiative.) • 1998 - . Continued research and commercial spin-off (which failed).

6. The Challenge A video sequence is awkward for information discovery: • Textual methods of information retrieval cannot be applied • Browsing requires the user to view the sequence. Fast skimming is difficult. • Computing requirements are demanding (MPEG-1 requires 1.2 Mbits/sec). Surrogates are required

7. Multi-Modal Information Discovery • The multi-modal approach to information retrieval • Computer programs to analyze video materials for clues • e.g., changes of scene • methods from artificial intelligence, e.g., speech recognition, natural language processing, image recognition. • analysis of video track, sound track, closed captioning if present, any other information. • Each mode gives imperfect information. Therefore use • many approaches and combine the evidence.

8. Informedia Library Creation Speech recognition Image extraction Natural language interpretation Segments with derived metadata Text Audio Video Segmentation

9. Informedia: Information Discovery Segments with derived metadata User Querying via natural language Browsing via multimedia surrogates Requested segments and metadata

10. Text Extraction Source Sound track: Automatic speech recognition using Sphinx II and III recognition systems. (Unrestricted vocabulary, speaker independent, multi-lingual, background sounds). Error rates 25% up. Closed captions: Digitally encoded text. (Not on all video. Often inaccurate.) Text on screen: Can be extracted by image recognition and optical character recognition. (Matches speaker with name.) Query Spoken query: Automatic speech recognition using the same system as is used to index the sound track. Typed by user

11. Image Understanding Informedia has developed specialized tools for various aspects of image understanding • scene break detection segmentation icon selection • image similarity matching • camera motion and object tracking • video-OCR (recognize text on screen) • face detection and association

12. Multimodal Metadata Extraction

13. An Evaluation Experiment Test corpus: • 602 news stories from CNN, etc. Average length 672 words. • Manually transcribed to obtained accurate text. • Speech recognition of text using Sphinx II (50.7% error rate) • Errors introduced artificially to give error rates from 0% to 80%. • Relative precision and recall (using a vector ranking) were used as measures of retrieval performance. As word error rate increased from 0% to 50%: • Relative precision fell from 80% to 65% • Relative recall fell from 90% to 80%

14. Speech recognition and retrieval performance

15. User Interface Concepts Users need a variety of ways to search and browse, depending on the task being carried out and preferred style of working • Visual icons one-line headlines film strip views video skims transcript following of audio track • Collages • Semantic zooming • Results set • Named faces • Skimming

17. Thumbnails, Filmstrips and Video Skims Thumbnail: • A single image that illustrates the content of a video Filmstrip: • A sequence of thumbnails that illustrate the flow of a video segment Video skim: • A short video that summarizes the contents of a longer sequence, by combining shorter sequences of video and sound that provide an overview of the full sequence

18. Creating a Filmstrip Separate video sequence into shots • Use techniques from image recognition to identify dramatic changes in scene. Frames with similar color characteristics are assumed to be part of a single shot. Choose a sample frame • Default is to select the middle frame from the shot. • If camera motion, select frame where motion ends. User feedback: • Frames are tied to time sequence.

19. Creating Video Skims Static: • Precomputed based on video and audio phrases • Fixed compression, e.g., one minute skim of 10 minute sequence Dynamic: • After a query, skim is created to emphasize context of the hit • Variable compression selected by user • Adjustable during playback

20. Limits to Scalability Informedia has demonstrated effective information discovery with moderately large collections Problems with increased scale: • Technical -- storage, bandwidth, etc. • Diversity of content -- difficult to tune heuristics • User interfaces -- complexity of browsing grows with scale

21. Lessons Learned • Searching and browsing must be considered integrated parts of a single information discovery process. • Data (content and metadata), computing systems (e.g., search engines), and user interfaces must be designed together. • Multi-modal methods compensate for incomplete or error-prone data.

22. CS 430 / INFO 430 Information Retrieval Lecture 23 Architecture of Information Retrieval Systems

23. Basic Architecture 1: Single Homogeneous Collection • Documents and indexes are held on a single computer system (may be several computers). • The user interface and search methods are selected for the specific service. Index Documents Examples: Medline (medical information) Cornell University library catalog

24. Basic Architecture 2: Several Similar Collections -- One Computer System • Several more or less similar collections are held on a single computer system. • Each collection is indexed separately using the same software, procedures, algorithms, etc. (but tuned for each collection, e.g., stoplists). • The user interface is the same (or very similar) for each service. Examples: OCLC's FirstSearch

25. Distributed Architecture 1: Standard Search Protocols Find x Strict adherence to standards allows any user interface to search any conforming search service. Find x

26. Distributed Architecture 2: Broadcast Search (a.k.a. Federated Search) Interface Service An interface server broadcasts a query to each collection, combines the results and returns them to the user. Examples: Dienst (digital library protocol), Web metasearch services Find x

27. Distributed Architecture 3: Centralized Search Services Search Service Batch indexing: Metadata about all items is accumulated in a central system. Real-time searching: The user (a) searches the central system, and (b) retrieves items from collections. Examples: Union catalogs, Web search services retrieve search Find x