The Experts below are selected from a list of 1896 Experts worldwide ranked by ideXlab platform
Rozi Fahrur - One of the best experts on this subject based on the ideXlab platform.
-
Document Grouping by Using Meronyms and Type-2 Fuzzy Association Rule Mining
'The Institute for Research and Community Services (LPPM) ITB', 2017Co-Authors: Rozi Fahrur, Sukmana FaridAbstract:The growth of the number of textual documents in the digital world, especially on the World Wide Web, is incredibly fast. This causes an accumulation of information, so we need efficient organization to manage textual documents. One way to accurately classify documents is using fuzzy association rules. The quality of the document clustering is affected by phase extraction of key terms and type of fuzzy logic system (FLS) used for clustering. The use of meronyms in the extraction of key terms to obtain cluster labels helps obtaining meaningful cluster labels and in addition ambiguities and uncertainties that occur in the rules of type-1 fuzzy logic systems can be overcome by using type-2 fuzzy sets. This study proposes a method of key term extraction based on meronyms with an initialization cluster using fuzzy association rule mining for document clustering. This method consists of four stages, i.e. preprocessing of the document, extraction of key terms with meronyms, extraction of candidate clusters, and cluster tree construction. Testing of this method was done with three different datasets: classic, Reuters, and 20 Newsgroup. Testing was done by comparing the overall F-measure of the method without meronyms and with meronyms. Based on the testing, the method with meronyms in the extraction of keywords produced an overall F-measure of 0.5753 for the classic dataset, 0.3984 for the Reuters dataset, and 0.6285 for the 20 Newsgroup dataset
-
Ekstraksi Kata Kunci Berdasarkan Hipernim Dengan Inisialisasi Klaster Menggunakan Fuzzy Association Rule Mining Pada Pengelompokan Dokumen
2015Co-Authors: Rozi FahrurAbstract:Pertumbuhan dunia digital dalam dokumen teks terutama di World Wide Web mengalami pertumbuhan pesat. Peningkatan dokumen teks ini menyebabkan terjadinya penumpukan informasi, sehingga diperlukan sebuah pengorganisasian yang efisien untuk pengelolaan dokumen teks. Salah satu metode yang dapat mengelompokkan dokumen dengan tepat adalah menggunakan fuzzy association rule. Tahap ekstraksi kata kunci serta tipe fuzzy yang digunakan berpengaruh terhadap kualitas pengelompokan dokumen. Penggunaan hipernim dalam ekstraksi kata kunci untuk mendapatkan suatu cluster label dapat memperluas makna dari cluster label, sehingga dapat diperoleh suatu meaningful cluster label, selain itu ambiguitas dan uncertainties yang terjadi di dalam aturan fuzzy logic systems (FLS) tipe-1 dapat diatasi dengan fuzzy set tipe-2. Penelitian ini mengusulkan sebuah metode yaitu ekstraksi kata kunci berdasarkan hipernim dengan inisialisasi klaster menggunakan fuzzy association rule mining pada pengelompokan dokumen. Metode ini terdiri dari empat tahap, yaitu : preprocessing dokumen, ekstraksi key terms dari hipernim, ekstraksi kandidat cluster, dan konstruksi cluster tree. Pengujian terhadap metode ini dilakukan dengan tiga jenis data berbeda, yaitu Classic, Reuters, dan 20 Newsgroup. Pengujian dilakukan dengan membandingkan nilai overall f-measure dari metode tanpa hipernim (level 0), metode dengan hipernim level 1, dan metode dengan hipernim level 2. Berdasarkan pengujian didapatkan bahwa penggunaan hipernim dalam ektraksi kata kunci mampu menghasilkan rata-rata overall f-measure sebesar 0.5783 untuk data classic, 0.4001 untuk data reuters, dan 0.5269 untuk data 20 Newsgroup. =========================================================================================================== The development of the digital world in text documents, especially on the World Wide Web is very fast. With the rapid growth of text document can lead to information overload, so we need an efficient method to manage of text documents. One of method that can accurately classify documents is using fuzzy association rule. Key terms extraction stage and type of fuzzy that used for clustering affected on the quality of the document clustering. Using hipernim in the key terms extraction to obtain a cluster label can expand the meaning of cluster label and obtain a meaningful cluster label, the ambiguities and uncertainties that occur in the rules of fuzzy logic systems (FLS) type-1 can be overcome with fuzzy sets type-2. This research propose a method of key terms extraction based on hipernim with initialization cluster using fuzzy association rule mining in document clustering. This method consists of four stages, that is: preprocessing documents, key terms extraction with hipernim, candidate clusters extraction, and cluster tree construction. This research method was tested on three different types of data, that is : Classic, Reuters, and 20 Newsgroup. We have conducted experiment by comparing overall f-measure of method without hypernym (level 0), method with hypernym level 1, and method with hypernym level 2. Based on testing, method with hypernym in the extraction of keyword can produce overall f-measure 0.5783 for classic data, 0.4001 for reuters data, and 0.5269 for 20 Newsgroup data
-
EKSTRAKSI KATA KUNCI BERDASARKAN HIPERNIM DENGAN INISIALISASI KLASTER MENGGUNAKAN FUZZY ASSOCIATION RULE MINING PADA PENGELOMPOKAN DOKUMEN
'Lembaga Penelitian dan Pengabdian kepada Masyarakat ITS', 2015Co-Authors: Rozi Fahrur, Fatichah Chastine, Purwitasari DianaAbstract:Pertumbuhan dunia digital dalam dokumen tekstual terutama di World Wide Web mengalami pertumbuhan pesat. Pen-ingkatan dokumen tekstual ini menyebabkan terjadinya penumpukan informasi, sehingga diperlukan sebuah pengorgan-isasian yang efisien untuk pengelolaan dokumen tekstual. Salah satu metode yang dapat mengelompokkan dokumen dengan tepat adalah menggunakan fuzzy association rule. Tahap ekstraksi kata kunci serta tipe fuzzy yang digunakan berpengaruh terhadap kualitas pengelompokan dokumen. Penggunaan hipernim dalam ekstraksi kata kunci untuk mendapatkan suatu klaster label dapat memperluas makna dari klaster label, sehingga dapat diperoleh suatu meaningful klaster label, selain itu ambiguitas dan uncertainties yang terjadi di dalam aturan fuzzy logic systems (FLS) tipe-1 dapat diatasi dengan fuzzy set tipe-2. Penelitian ini mengusulkan sebuah metode yaitu ekstraksi kata kunci berdasarkan hipernim dengan inisialisasi klaster menggunakan fuzzy association rule mining pada pengelompokan dokumen. Metode ini terdiri dari empat tahap, yaitu : preprocessing dokumen, ekstraksi key terms dari hipernim, ekstraksi kandidat klaster, dan konstruksi klaster tree. Pengujian terhadap metode ini dilakukan dengan tiga jenis data berbeda, yaitu Classic, Reuters, dan 20 Newsgroup. Pengujian dilakukan dengan membandingkan nilai overall f-measure dari metode tanpa hipernim (level 0), hipernim level 1, dan hipernim level 2. Berdasarkan pengujian didapatkan bahwa penggunaan hipernim dalam ektraksi kata kunci mampu menghasilkan rata-rata overall f-measure sebesar 0.5783 untuk data classic, 0.4001 untuk data reuters, dan 0.5269 untuk data 20 Newsgroup
Purwitasari Diana - One of the best experts on this subject based on the ideXlab platform.
-
EKSTRAKSI KATA KUNCI BERDASARKAN HIPERNIM DENGAN INISIALISASI KLASTER MENGGUNAKAN FUZZY ASSOCIATION RULE MINING PADA PENGELOMPOKAN DOKUMEN
'Lembaga Penelitian dan Pengabdian kepada Masyarakat ITS', 2015Co-Authors: Rozi Fahrur, Fatichah Chastine, Purwitasari DianaAbstract:Pertumbuhan dunia digital dalam dokumen tekstual terutama di World Wide Web mengalami pertumbuhan pesat. Pen-ingkatan dokumen tekstual ini menyebabkan terjadinya penumpukan informasi, sehingga diperlukan sebuah pengorgan-isasian yang efisien untuk pengelolaan dokumen tekstual. Salah satu metode yang dapat mengelompokkan dokumen dengan tepat adalah menggunakan fuzzy association rule. Tahap ekstraksi kata kunci serta tipe fuzzy yang digunakan berpengaruh terhadap kualitas pengelompokan dokumen. Penggunaan hipernim dalam ekstraksi kata kunci untuk mendapatkan suatu klaster label dapat memperluas makna dari klaster label, sehingga dapat diperoleh suatu meaningful klaster label, selain itu ambiguitas dan uncertainties yang terjadi di dalam aturan fuzzy logic systems (FLS) tipe-1 dapat diatasi dengan fuzzy set tipe-2. Penelitian ini mengusulkan sebuah metode yaitu ekstraksi kata kunci berdasarkan hipernim dengan inisialisasi klaster menggunakan fuzzy association rule mining pada pengelompokan dokumen. Metode ini terdiri dari empat tahap, yaitu : preprocessing dokumen, ekstraksi key terms dari hipernim, ekstraksi kandidat klaster, dan konstruksi klaster tree. Pengujian terhadap metode ini dilakukan dengan tiga jenis data berbeda, yaitu Classic, Reuters, dan 20 Newsgroup. Pengujian dilakukan dengan membandingkan nilai overall f-measure dari metode tanpa hipernim (level 0), hipernim level 1, dan hipernim level 2. Berdasarkan pengujian didapatkan bahwa penggunaan hipernim dalam ektraksi kata kunci mampu menghasilkan rata-rata overall f-measure sebesar 0.5783 untuk data classic, 0.4001 untuk data reuters, dan 0.5269 untuk data 20 Newsgroup
John A. Bargh - One of the best experts on this subject based on the ideXlab platform.
-
Coming Out in the Age of the Intemet: Identity "Demarginalization" Through Virtual Group Participation
2014Co-Authors: Katelyn Y. A. Mckenna, John A. BarghAbstract:Internet Newsgroups allow individuals to interact with others in a relatively anonymous fashion and thereby provide individuals with concealable stigmatized identities a place to belong not otherwise available. Thus, membership in these groups should become an important part of identity. Study 1 found that members of Newsgroups dealing with marginalized-concealable identifies modified their Newsgroup behavior on the basis of reactions of other members, unlike members of marginalized-conspicuous ormainstream Newsgroups. This increase in identity importance from Newsgroup artici-pation was shown in both Study 2 (marginalized sexual identifies) and Study 3 (marginalized ideological identities) to lead to greater self-acceptance, as well as coming out about the secret identity to family and friends. Results supported the view that Internet groups obey general principles of social group functioning and have real-life consequences for the individual. I just thought, "Oh God. What if they pick up that I'm gay? " It was that fear and shame.... I watched the whole Gay Pride march in Washington in 1993, and I wept when I saw that, I mean I cried so hard, thinking "I wish I could be there, " because I never felt like I belonged anywhere.--Ellen DeGeneres, Time magazine intervie
-
coming out in the age of the internet identity demarginalization through virtual group participation
Journal of Personality and Social Psychology, 1998Co-Authors: Katelyn Y. A. Mckenna, John A. BarghAbstract:Internet Newsgroups allow individuals to interact with others in a relatively anonymous fashion and thereby provide individuals with concealable stigmatized identities a place to belong not otherwise available. Thus, membership in these groups should become an important part of identity. Study 1 fou
Howard T Welser - One of the best experts on this subject based on the ideXlab platform.
-
You Are Who You Talk To:
2008Co-Authors: Danyel Fisher, Marc Smith, Howard T WelserAbstract:Understanding the social roles of the members a group can help to understand the social context of the group. We present a method of applying social network analysis to support the task of characterizing authors in Usenet Newsgroups. We compute and visualize networks created by patterns of replies for each author in selected Newsgroups and find that second-degree ego-centric networks give us clear distinctions between different types of authors and Newsgroups
-
visualizing the signatures of social roles in online discussion groups
Journal of Social Structure, 2007Co-Authors: Howard T Welser, Eric Gleave, Danyel FisherAbstract:Social roles in online discussion forums can be described by patterned characteristics of communication between network members which we conceive of as 'structural signatures.' This paper uses visualization methods to reveal these structural signatures and regression analysis to confirm the relationship between these signatures and their associated roles in Usenet Newsgroups. Our analysis focuses on distinguishing the signatures of one role from others, the role of "answer people." Answer people are individuals whose dominant behavior is to respond to questions posed by other users. We found that answer people predominantly contribute one or a few messages to discussions initiated by others, are disproportionately tied to relative isolates, have few intense ties and have few triangles in their local networks. OLS regression shows that these signatures are strongly correlated with role behavior and, in combination, provide a strongly predictive model for identifying role behavior (R =.72). To conclude, we consider strategies for further improving the identification of role behavior in online discussion settings and consider how the development of a taxonomy of author types could be extended to a taxonomy of Newsgroups in particular and discussion systems in general. 2
-
Visualizing the signatures of social roles in online discussion groups, The
2007Co-Authors: Howard T Welser, Danyel Fisher, Eric Gleave, Marc SmithAbstract:Abstract: Social roles in online discussion forums can be described by patterned characteristics of communication between network members which we conceive of as ‘structural signatures. ' This paper uses visualization methods to reveal these structural signatures and regression analysis to confirm the relationship between these signatures and their associated roles in Usenet Newsgroups. Our analysis focuses on distinguishing the signatures of one role from others, the role of “answer people. " Answer people are individuals whose dominant behavior is to respond to questions posed by other users. We found that answer people predominantly contribute one or a few messages to discussions initiated by others, are disproportionately tied to relative isolates, have few intense ties and have few triangles in their local networks. OLS regression shows that these signatures are strongly correlated with role behavior and, in combination, provide a strongly predictive model for identifying role 2 behavior (R =.72). To conclude, we consider strategies for further improving the identification of role behavior in online discussion settings and consider how the development of a taxonomy of author types could be extended to a taxonomy of Newsgroups in particular and discussion systems in general
-
you are who you talk to detecting roles in usenet Newsgroups
Hawaii International Conference on System Sciences, 2006Co-Authors: Danyel Fisher, Howard T WelserAbstract:Understanding the social roles of the members a group can help to understand the social context of the group. We present a method of applying social network analysis to support the task of characterizing authors in Usenet Newsgroups. We compute and visualize networks created by patterns of replies for each author in selected Newsgroups and find that second-degree ego-centric networks give us clear distinctions between different types of authors and Newsgroups. Results show that Newsgroups vary in terms of the populations of participants and the roles that they play. Newsgroups can be characterized by populations that include question and answer Newsgroups, conversational Newsgroups, social support Newsgroups, and flame Newsgroups. This approach has applications for both researchers seeking to characterize different types of social cyberspaces as well as participants seeking to distinguish interaction partners and content authors.
-
picturing usenet mapping computer mediated collective action
Journal of Computer-Mediated Communication, 2005Co-Authors: Tammara Combs Turner, Danyel Fisher, Marc A. Smith, Howard T WelserAbstract:Usenet is a complex socio-technical phenomenon, containing vast quantities of information. The sheer scope and complexity make it a challenge to understand the many dimensions across which people and communication are interlinked. In this work, we present visualizations of several aspects and scales of Usenet that combine to highlight the range of variation found in Newsgroups. We examine variations within hierarchies, Newsgroups, authors, and social networks. We find a remarkable diversity, with clear variations that mark starting points for mapping the broad sweep of behavior found in this and other social cyberspaces. Our findings provide the basis for initial recommendations for those cultivating, managing, contributing, or consuming collectively constructed conversational content.
Marc A. Smith - One of the best experts on this subject based on the ideXlab platform.
-
1 Grand Central Usenet: The Design and Evaluation of a Thread-Based Usenet Browser
2015Co-Authors: Carman Neustaedter, Marc A. Smith, Gina Danielle VenoliaAbstract:Interfaces to online discussion spaces, such as email discussions, lists, and Newsgroups, do a poor job of representing the structure and temporal development of conversation threads. These limitations contribute to user overload and to the erosion of the value of these channels. In this paper, we present an alternative interface to threaded conversations, Grand Central Usenet, which features a graphical interface component that highlights the size, structure, and development of conversation threads. We harnessed this interface to Usenet Newsgroup data and conducted a user study that contrasted this interface with a standard message browsing tool. Users showed significant improvements in productivity, reports of ease-of-use, and satisfaction with our design in contrast to a widely used standard interface
-
picturing usenet mapping computer mediated collective action
Journal of Computer-Mediated Communication, 2005Co-Authors: Tammara Combs Turner, Danyel Fisher, Marc A. Smith, Howard T WelserAbstract:Usenet is a complex socio-technical phenomenon, containing vast quantities of information. The sheer scope and complexity make it a challenge to understand the many dimensions across which people and communication are interlinked. In this work, we present visualizations of several aspects and scales of Usenet that combine to highlight the range of variation found in Newsgroups. We examine variations within hierarchies, Newsgroups, authors, and social networks. We find a remarkable diversity, with clear variations that mark starting points for mapping the broad sweep of behavior found in this and other social cyberspaces. Our findings provide the basis for initial recommendations for those cultivating, managing, contributing, or consuming collectively constructed conversational content.
-
assessing differential usage of usenet social accounting meta data
Human Factors in Computing Systems, 2005Co-Authors: A Bernheim J Brush, Tammara Combs Turner, Xiaoqing Wang, Marc A. SmithAbstract:We describe a usage study of NetscanTech, a system that generates and publishes daily a range of social metrics across three dimensions: Newsgroup, author, and thread, for a set of approximately 15,000 technical Newsgroups in Usenet. We bring together three interlinked datasets: survey data, usage log data and social accounting data from Usenet participation, to triangulate the relationship between various user roles and differential usage of social metrics in NetscanTech. We found our most frequent users focused on information related to individual authors far more than any other information provided. In contrast, users that visited less frequently focused more on information related to Newsgroups and viewing Newsgroup metrics. Our results suggest features that designers and developers of online communities may wish to include in their interfaces to support the cultivation of different community roles.
-
inhabitant s uses and reactions to usenet social accounting data
2004Co-Authors: Byron Burkhalter, Marc A. SmithAbstract:Netscan social accounting data is applied both to Newsgroups and participants to distinguish between useful Newsgroups and those that are noisy or fractious and between authors who may be regulars or interlopers, nice people or not so nice, and those authors who have quality responses as well as a quantity of responses. Through these and other structural measures Newsgroups and their members are typified, categorised and understood from a perspective not possible (or excessively costly to manually construct) through a simple archive of messages. Social accounting tools present the historical and sociological tracks of the Newsgroup and are used to perform functions that seem extremely similar to those used by offline groups and organisations. Social accounting data is not merely useful for our understanding of Usenet Newsgroups but may be becoming a vital and commonly used tool by the members of these kinds of discussion spaces themselves.
-
Visualization Components for Persistent Conversation
2001Co-Authors: Marc A. Smith, Andrew FioreAbstract:An appropriately designed interface to persistent, threaded conversations could reinforce socially beneficial behavior by prominently featuring how frequently and to what degree each user exhibits such behaviors. Based on the data generated by the Netscan data-mining project [9], we have developed a set of tools for illustrating the structure of discussion threads like those found in Usenet Newsgroups and the patterns of participation within the discussions. We describe the benefits and challenges of integrating these tools into a multi-faceted dashboard for navigating and reading discussions in social cyberspaces like Usenet and related interaction media. Visualizations of the structure of online discussions have applications for research into the sociology of online groups as well as possible interface designs for their members. Keywords Visualization, persistent conversation, asynchronous threaded discussions, Usenet, Newsgroup, social cyberspace
