United States Patent  [ii] Patent Number: 5,727,129
Barrett et al.  Date of Patent: Mar. 10, 1998
 NETWORK SYSTEM FOR PROFILING AND ACTIVELY FACILITATING USER ACTIVITIES
 Inventors: Robert Carl Barrett; Daniel Clark Kellem. both of San Jose; Paul Philip Maglio, Santa Cruz, all of Calif.
 Assignee: International Business Machines
Corporation. Armonk. N.Y.
 Appl. No.: 659,100
 Filed: Jun. 4,1996
 Int CI.6 G06F 3/00
 U.S. CI 395/12; 395/357; 395/353;
 Field of Search 395/12. 1. 10.
395/326-358. 200.01, 200.02. 200.11, 200.08.
 References Cited
U.S. PATENT DOCUMENTS
4,931.950 6/1990 Meet al 395/12 X
5,103,498 4/1992 Lanier et al 395/12 X
5,204,947 4/1993 Bernstein et al 395/157
5,208,745 5/1993 Quentin et al 395/12 X
5^39,617 8/1993 Gardner et al 395/12
5,297,249 3/1994 Bernstein et al 395/156
5,333,237 7/1994 Stefanopoulos et al 395/12
5,355,472 10/1994 Lewis 395/600
5,390281 2/1995 Luciw et al 395/12
5,506,937 4/1996 Ford et al 395/12
5,560,011 9/1996 Uyama 395/12 X
O'Leary, "Al and Navigation on the Internet and Intranet", IEEE Expert, pp. 8-10. Apr. 1996.
Chang et al.. "Intelligent Database Retrieval By Visual Reasoning". IEEE, pp. 459-464. 1990.
Passani et al.. "Learning from Hotlists and Coldlists: Towards a WWW information Filtering and Seeking Agent", IEEE, pp. 492^95, 1995.
Story et al.. "The RightPages Image-Based Electronic Library for Altering and Browsing", IEEE, pp. 17-26,1992.
M. Balabanovic & Y. Shoham. Learning Information Retrieval Agents: Experiments with Automated Web Browsing, Dept. of Computer Science, Stanford Univ., California pp. 13-17 (email@example.com).
R. Armstrong. D. Freitag. T. Joachims, & T. Mitchell, WebWatcher: A Learning Apprentice for the World Wide Web, School of Computer Science, Carnegie Mellon Univ. 1/20/95, pp. 6-12. AAAI Spring Symposum. Mar. 27-29. 1995. Stanford Univ. Info Gathering for Heterogeneous. Distributed Environment.
Primary Examiner—John E. Breene
Attorney, Agent, or Firm—James C. Pintner
A system and method are provided for use with an communication and information network, such at the Internet World Wide Web, for assisting a user in accessing information stored at remote network sites based on the user's past history of network usage. An archive is maintained of remote sites accessed and instances in which the same remote sites are accessed in sequence. Statistics regarding information such as the number of time a site has been accessed, and the times a given set of sites have been accessed in sequence, are maintained. This information may be displayed upon command. Based on this information, information items are identified which the user is predicted to be likely to want to access. This information is predownloaded. without express user command, so that if the user does enter a command, the response time is advantageously fast.
22 Claims, 11 Drawing Sheets