K. Lang. News Weeder: Learning to filter netnews. In Proc. 12th Int'l Conf. Machine Learning, San Francisco, 1995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....value for 300 documents (k = 3, f = 30) 20 40 60 80 100 120 140 160 180 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 #clusters(k) Figure 6. The percentage increase in objective function value of our refinement algorithm over spherical k means on the 20 newsgroup data set (f = 20) groups [12]. Some of the newsgroups are closely related to each other (e.g. comp.graphics, comp.os.ms windows.misc and comp.windows.x) while others are unrelated (e.g. alt.atheism, rec.sport.baseball and sci.space) The headers for each of the messages were removed so that they do not bias clustering ....

K. Lang. News Weeder: Learning to filter netnews. In Proc. 12th Int'l Conf. Machine Learning, San Francisco, 1995.

....way. If Pl, Pn are the individual probabilities, and I is the set of indexes for which a bit is set in a given test vector, then the probability that the corresponding link was followed is determined by 1 1 [iei(1 Pi) TFIDF with cosine similarity measure [Salton and McGill, 1983; Lang, 1995] is a method de veloped in information retrieval. In the gen eral case at first a vector V of words is cre ated. In our experiments it is already given by the representation described above. Every instance can now be represented as a vector with the same length as V, replacing every word by a ....

K. Lang, News Weeder: Learning to Filter Netnews, to be submitted to the In- ternational Conference on Machine Learn- ing, 1995

.... 1990 ] in detecting credit card fraud [ Stolfo et al. 1997a ] in steering vehicles driving autonomously on public highways at 70 miles an hour [ Pomerleau, 1992 ] in predicting stock option pricing [ Malliaris Salchenberger, 1993 ] and in computing customized electronic newspapers [ K.Lang, 1995 ] to name a few applications. Many large business institutions and market analysis firms attempt to distinguish the low risk (high profit) potential customers by learning simple categorical classifications of their potential customer base. Similarly, defense and intelligence operations utilize ....

K.Lang. 1995. NEWS WEEDER: Learning to filter net news. In Prieditis, A., and S.Russel., eds., Proc. 12th Intl. Conf. Machine Learning, 331--339. Morgan Kaufmann.

....relatives of the proposed application exist in the form of various web agents and recommendation systems. Those applications use text analysis (content based approach) and user matching techniques (collaborative approach) to assist users browsing the web [2,3,4,5,6,7,8,9,10] filter usenet news [11,12] or find specialized info on the internet [13,14,15,16] A well known recommendation system can be found at Amazon.com. There, users create their own profiles, which record the items they looked at or bought. Amazon recommends by looking at other users who bought the same items, and suggests ....

K. Lang, "News Weeder: Learning to Filter Netnews," Proc. 12th Int'l Conf. Machine Learning,Morgan Kaufmann, San Francisco, 1995, pp. 331--339.

....1.1. 2 Approaches to Information filtering Information filtering techniques have been applied to several application areas including scientific publications [Luh58] technical memos and reports [FD92, FC97] Usenet news [FS91, Bac91, SK92, Ste92b, Bac92, JH92, SM93, Mae94, KHL 94, RIS 94, MS94, Lan95, YG95, Moc96, MRK97, MRK 97] electronic mail [Mye80, MGT 87, Pol88, GNOT92, Ter91, Ter93, LMM94] books [Ric79b, MR98] application program know how [LN98] the finding of experts [SW93, KSS96] Web pages [RM96, HT96, Bal97, THA 97, PB97, RP97, Bie98] classified ads [GGKS95] movies [Kay95, ....

....processes, e.g. the user gaining experience or growing older. An example of a gradual interest change is a user s taste in music as it changes while the user gets older. Most authors who refer to interest changes only in general, implicitly refer to gradual interest changes (e.g. Luh58, Bac91, Lan95] Abrupt interest changes, sometimes also referred to as interest shifts [LMMP96] have been recognized as happening induced by events [Mar95] Such changes may occur, for example, in a professional environment due to a change in the job assignment or the initiation of a new job [LMMP96] As a ....

[Article contains additional citation context not shown here]

K. Lang. News Weeder: Learning to filter Netnews. In Proceedings of the International Conference on Machine Learning, Tahoe City, CA., 1995.

....of objects for each category is a much studied topic in the statistics, machine learning, and data mining literature. See, for instance, 13] for a comprehensive review of various classification techniques. Naive Bayes classifiers [7] are competitive with other techniques in accuracy [3] 10] [8] [15] 12] They are also fast: building the model requires only a single pass over the documents and they quickly classify new documents. Our proposed solution is also Bayesian. The observation that the classification techniques can be used to assign documents to a hierarchy has been previously ....

K. Lang. News Weeder: Learning to Filter Net-News. In Proc. of the 12th Int'l Conf. on Machine Learning, pages 331--339, 1995.

....et al. 2] developed WebWatcher, a system assisting users seeking information on the Web (see also [32] Newsgroup related problems are investigated in [30] that proposes a method for Automatic E mail classi cation. Krulwich and Burkey [17] address newsgroup message categorization and Lang [18] proposes to learn user interests and lter news messages accordingly. Other authors write about similar problems (see e.g. 27] The author identi cation problem was previously addressed by a variety of statistical techniques with various document representations. Those include: character ....

K. Lang. News Weeder: Learning to Filter Netnews. In Proc. of the 12th International Conference on Machine Learning ICML95, 1995.

.... disease diagnosis [61] in predicting glucose levels for diabetic patients [22] in detecting credit card fraud [65] in steering vehicles driving autonomously on public highways at 70 miles an hour [50] in predicting stock option pricing [46] and in computing customized electronic newspapers[33], to name a few applications. Many large business institutions and market analysis firms attempt to distinguish the low risk (high profit) potential customers by learning simple categorical classifications of their potential customer base. Similarly, defense and intelligence operations utilize ....

K.Lang. News weeder: Learning to filter net news. In A.Prieditis and S.Russel, editors, Proc. 12th Intl. Conf. Machine Learning, pages 331--339. Morgan Kaufmann, 1995.

....for hierarchy reorganization, document routing, and identification of misfiled documents. We decided to base our classifier on the Naive Bayes model [Goo65] for the following reasons: Naive Bayes classifiers are very competitive with other techniques for text classification [CDAR97] LR94] [Lan95] [PB97] MN98] 2 They stabilize quickly [Koh96] which supports automated hierarchy reorganization with a limited number of examples. They are fast. They can be constructed quickly with a single pass over the documents, making them suitable for on line model creation; they also quickly ....

K. Lang. News Weeder: Learning to Filter Net-News. In Proc. of the 12th Int'l Conf. on Machine Learning, pages 331--339, 1995.

....document routing Inbox visualization, and identification of misfiled documents. We decided to base our classifier on the Naive Bayes model [Goo65] for the following reasons: ffl Naive Bayes classifiers are very competitive with other techniques for text classification [CDAR97] LR94] [Lan95] [PB97] MN98] 2 ffl They stabilize quickly [Koh96] which is useful for hierarchy reorganization. ffl They are fast. They can be constructed quickly with a single pass over the documents, making them suitable for on line model creation; they also quickly classify incoming documents [CDAR97] ....

K. Lang. News Weeder: Learning to Filter Net-News. In Proc. of the 12th Int'l Conf. on Machine Learning, pages 331--339, 1995.

....browsing WWW Lieberman [9] Letizia MIT browsing WWW Mladeni c [4] Personal WebWatcher CMU, IJS browsing WWW Pazzani et al. 10, 11] Syskill Webert UCI browsing WWW Shavlik et al. 12] WAWA Wisconsin Uni. browsing WWW Bollacker et al. 13] CiteSeer TX, NEC corp. finding papers UMIACS on WWW Lang [14] NewsWeeder CMU Usenet news filtering Krulwich et al. 15] ContactFinder Andersen Consult. finding expert Burke et al. 16, 17] FAQFinder Uni. of Chicago answer question Kamba et al. 18] Antagonomy NEC corp. personalized newspaper LaMacchia [19] Internet Fish MIT find info. on Internet Mitchell ....

....publications. The system gets keywords from the user and calls search engines to find relevant papers (relevant PostScript files on the Web) and extracts header, abstract, citations from the papers. It also enables finding of similar papers based on the common citations in the papers. Lang [14] developed NewsWeeder, a system for electronic news filtering that uses text classification to generate a model of user s interests. The system uses a World Wide Web interface to enable user to access the news in a usual way and to enable the system to collect user s ratings as feedback. ....

[Article contains additional citation context not shown here]

Lang, K., News Weeder: Learning to Filter Netnews, Proc. of the 12th International Conference on Machine Learning ICML95, 1995.

.... heart disease diagnosis [29] in predicting glucose levels for diabetic patients [12] in detecting credit card fraud [30] in steering vehicles driving autonomously on public highways at 70 miles an hour [24] in predicting stock option pricing [23] in computing customizing electronic newspapers[15] etc. Many large business institutions and market analysis firms attempt to distinguish the low risk (high profit) potential customers by learn simple categorical classifications of their potential customer data base. Similarly, defense and intelligence operations utilize similar methodologies on ....

K.Lang. News weeder: Learning to filter net news. In A.Prieditis and S.Russel, editors, Proc. 12th Intl. Conf. Machine Learning, pages 331--339. Morgan Kaufmann, 1995.

.... heart disease diagnosis [39] in predicting glucose levels for diabetic patients [17] in detecting credit card fraud [41] in steering vehicles driving autonomously on public highways at 70 miles an hour [33] in predicting stock option pricing [32] in computing customizing electronic newspapers[21] etc. Many large business institutions and market analysis firms attempt to distinguish the low risk (high profit) potential customers by learn simple categorical classifications of their potential customer data base. Similarly, defense and intelligence operations utilize similar methodologies on ....

K.Lang. News weeder: Learning to filter net news. In A.Prieditis and S.Russel, editors, Proc. 12th Intl. Conf. Machine Learning, pages 331--339. Morgan Kaufmann, 1995.

.... disease diagnosis [53] in predicting glucose levels for diabetic patients [20] in detecting credit card fraud [57] in steering vehicles driving autonomously on public highways at 70 miles an hour [43] in predicting stock option pricing [40] and in computing customized electronic newspapers[27], to name a few applications. Many large business institutions and market analysis firms attempt to distinguish the low risk (high profit) potential customers by learning simple categorical classifications of their potential customer base. Similarly, defense and intelligence operations utilize ....

K.Lang. News weeder: Learning to filter net news. In A.Prieditis and S.Russel, editors, Proc. 12th Intl. Conf. Machine Learning, pages 331--339. Morgan Kaufmann, 1995.

....Web pages from the user and learns a user profile from them. Pages are separated according to their topic and a separate profile is learned for each topic. Generated user profile is used to form queries for the existing search engines in order to get more potentially interesting documents. Lang [31] developed NewsWeeder, a system for electronic news filtering that uses text learning to generate model of user interests. The system uses World Wide Web interface to enable user to access the news in a usual way and to enable the system to collect user s ratings as feedback. NewsWeeder ....

....is to help user Content based agents Armstrong et al. 3] WebWatcher CMU browsing WWW Balabanovi c and Shoham [4] Lira Stanford browsing WWW Goldman et al. 16] Musag Hebrew Univ. browsing WWW Lieberman [33] Letizia MIT browsing WWW Pazzani et al. 40, 41] Syskill Webert UCI browsing WWW Lang [31] NewsWeeder CMU Usenet news filtering Krulwich and Burkey [27] ContactFinder Andersen Consult. finding expert Burke et al. 9, 10] FAQFinder Univ. of Chicago answer question Kamba et al. 22] Antagonomy NEC corp. personalized newspaper LaMacchia [30] Internet Fish MIT extract info. from Internet ....

[Article contains additional citation context not shown here]

Lang, K., News Weeder: Learning to Filter Netnews, Proc. of the 12th International Conference on Machine Learning ICML95, 1995.

No context found.

K. Lang. News Weeder: Learning to filter netnews. In Proc. 12th Int'l Conf. Machine Learning, San Francisco, 1995.

No context found.

K. Lang. News Weeder: Learning to filter netnews. In Proc. 12th Int'l Conf. Machine Learning, San Francisco, 1995.

No context found.

K. Lang. News Weeder: Learning to filter netnews. In Proc. 12th Int'l Conf. Machine Learning, pages 331--339, 1995.

No context found.

K. Lang. News Weeder: Learning to filter netnews. In Proc. 12th Int'l Conf. Machine Learning, pages 331--339, San Francisco, 1995.

No context found.

K. Lang, "News Weeder: Learning to Filter Netnews, Proc. 12th Int'l Conf. Machine Learning (ICML 95), Morgan Kaufmann, San Francisco, 1995, pp. 313--339.

No context found.

Itell. 12:993--1001. K.Lang. 1995. News weeder: Learning to filter net news. In A.Prieditis, and S.Russel., eds., Proc. 12th Intl. Conf. Machine Learning, 331--339. Morgan Kaufmann.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC