Showing posts with label data-miner. Show all posts
Showing posts with label data-miner. Show all posts

Thursday, July 23, 2009

Call for Papers IIJ Special Issue on Data Mining

***********************************************************
Call for Papers IIJ Special Issue on Data Mining

The Internetworking Indonesia Journal (IIJ)
Indonesian Journal of ICT and Internet Development
(http://www.InternetworkingIndonesia.org)
***********************************************************

Deadline extended to 30 September 2009


Introduction to the Special Issue
The information age today is characterized by information overload, something which is clearly evident everywhere within the digital society. The form of information varies greatly, ranging from structured transactional data and semi structured web pages to unstructured multimedia data. Data mining is an important field since it helps us analyze that information at a higher rate than our natural abilities, thereby aiding us in absorbing that information in a meaningful fashion. This Special Issue on Data Mining aims to cover a broad range of topics, and solicits research paper submissions in the broad field of data mining. Topics of interest include – but are not limited to – the following:

• Text Mining
• Web Mining
• Data preprocessing
• Visual mining
• Multimedia data mining
• Mining frequent pattern
• Classification and prediction
• Cluster analysis
• Stream data mining
• Sequence data mining
• Graph mining

Special Issue dates:
Submission Extended Deadline: 30 September 2009
Acceptance Notification: 14 November 2009
Camera‐ready due: 12 December 2009

Guest Editors
The Special Issue solicits Research Papers in the area of data mining, though other papers are welcome for the future issues of the journal. Submission and correspondence for the special issue should be made via email to either the Guest Editors or the journal editors. Submissions will be reviewed anonymous and externally.
• Moch Arif Bijaksana, M Tech, Data Mining Center, Telkom Institute of Technology (IT Telkom) Email: arifbijaksana@gmail.com
• Dr. Anto Satriyo Nugroho, Center for Information & Communication Technology, Agency for the Assessment & Application of Technology (PTIK‐BPPT) Email: asnugroho@gmail.com

= =

About the IIJ
The Internetworking Indonesia Journal is a semi‐annual electronic journal devoted to the timely study of the Information and Communication Technology (ICT) and Internet development in Indonesia. The journal seeks highquality manuscripts on the challenges and opportunities presented by information technology and the Internet in Indonesia. The IIJ specially encourages submission by postgraduate students in the field of ICT.

Focus & Scope
The Internetworking Indonesia Journal aims to become the foremost publication for practitioners, teachers, researchers and policy makers to share their knowledge and experience in the design, development, implementation, and the management of ICT and the Internet in Indonesia.

Submission Guidelines
The journal accepts a variety of manuscripts in either English or Bahasa Indonesia. Please review the descriptions below and identify the submission type best suited to your intended submission:
• Research Papers: Research papers report on results emanating from research projects, both theoretical and practical in nature.
• Short papers: Short research papers provide an introduction to new developments or advances regarding on‐going work.
• Policy Viewpoint: Policy Viewpoints explore competing perspectives in the Indonesian policy debate that are informed by academic research.
• Teaching Innovation: Teaching Innovation papers explore creative uses of information technology tools and the Internet to improve learning and education in Indonesia.
• Book Reviews: A review of a book, or other book‐length document, such as a government report or foundation report. (3 pages max.)

Manuscript Language
The Internetworking Indonesia Journal accepts and publishes papers in Bahasa Indonesia and English.

Manuscript Submission
Manuscripts should be submitted according to the IEEE Guide for authors, and will be refereed in the standard way. Manuscript pages should not exceed 7 pages of the IEEE 2‐column format preferably in a MS‐Word file format. Links to the IEEE style files can be found at www.InternetworkingIndonesia.org under Submissions.
Manuscripts submitted to the journal must not have been previously published or committed to another publisher under a copyright transfer agreement, and must not be under consideration by another journal. Authors of accepted papers are responsible for the Camera Ready Copy (CRC) in the IEEE 2‐column format (MS‐Word file).
Authors are advised that no revisions of the manuscript can be made after acceptance by the Editor for publication.
The benefits of this procedure are many, with speed and accuracy being the most obvious. We look forward to working with your electronic submission which will allow us to serve you more efficiently.

Co‐Editors
Thomas Hardjono, PhD (MIT Kerberos Consortium, MIT, USA)
Budi Rahardjo, PhD (ITB, Indonesia)
Kuncoro Wastuwibowo, MSc (PT. Telkom Indonesia)

Editorial Advisory Board
Prof. Edy Tri Baskoro (ITB, Indonesia)
Mark Baugher, MA (Cisco Systems, USA)
Lakshminath Dondeti, PhD (Qualcomm, USA)
Prof. Svein Johan Knapskog, PhD (NTNU, Norway)
Prof. Merlyna Lim, PhD (Arizona State University, USA)
Prof. Bambang Parmanto, PhD (University of Pittsburgh, USA)
Prof. Wishnu Prasetya, PhD (Utrecht University, The Netherlands)
Prof. Jennifer Seberry, PhD (University of Wollongong, Australia)
Prof. Willy Susilo, PhD (University of Wollongong, Australia)
Prof. David Taniar, PhD (Monash University, Australia)

Open Access Publication Policy
The Internetworking Indonesia journal provides open access to all of its content on the principle that making research freely available to the public supports a greater global exchange of knowledge. This follows the philosophy of the Open Journal Systems (see the Public Knowledge Project at pkp.sfu.ca). The journal will be published electronically and there are no subscription fees. Such access is associated with increased readership and increased citation of an author's work.

Sunday, December 14, 2008

tulisan pelabelan peran semantik secara manual

Buat tulisan tentang pelabelan peran semantik bahasa Indonesia secara manual dengan mengacu ke FrameNet dan PropBank

Friday, December 12, 2008

BNC, ANC

British National Corpus (BNC) dan American National Corpus (ANC)

Indonesian?
Melayu?

Thursday, December 11, 2008

diskusi summarization

Setiap Kamis pagi dan Selsa sore saya ketemu dengan tim penelitian text summarization (Nur, Febi, Husnul, Devi dan kemungkinan sebentar lagi bertambah beberapa mahasiswa). Kosentrasi utama sekarang adalah "graph-based for ranking summarization" Ide dasarnya adalah menerapkan PageRank untuk summarization. pelopor teknik ini adlaah LexRank dan TextRank.

Diskusi tadi pai al. membahas.

* apa perbedaan LexRank dengan TextRank?

Pada paper LexRank ada tida teknik: pertama yang sederhana, kedua menggunakan PageRank, ketiga continuos. Yang paling utama adalah yang kedua. Dari konsep yang disampaikan pada tulisan LexRank dan pada tulisan TextRank pebedaanya hanyalah: LexRank untuk multi-document dan Textrank untuk single-document. Pada implementasinya untuk mengatasi tantangan informasi ganda untuk multi-document, LexRank menerapkan juga "reranker" setalah dilakukan perankingan menggunakan "PageRank-based". Kemudian juga pada paper LexRank ada tambahan informasi bagaimana PageRank direpresentasikan menggunakan matriks. Dengan representasi matriks operasi dan implementasinya bisa terbantu/lebih mudah.

* beberapa kemungkinan untuk memperbaiki akurasi LexRank dan TextRank.

* timestamp graph.

* evaluasi menggunakan Rouge.

Wednesday, December 10, 2008

sentiment classification

Setiap Rabu sore (pk.15.30an s/d 17.00) saya berdiskusi dengan Agung Pradeka, tentang "sentiment analysis"atau lebih tepatnya "sentiment classifcation".

Sentiment=perasaan, emosi. Terkait erat dengan itu adalah "opini/pendapat"

Salah satu pelopor awal penelitian tentang klasifikasi sentimen adalah Bo Pang et. al. (Bo Pang, Lillian Lee, Shivakumar Vaithyanathan: Thumbs up? Sentiment Classification using Machine Learning Techniques. CoRR cs.CL/0205070, 2002). Dalam tulisan ini mereka menyampaikan pentingnya klasifikasi sentimen. Dikatakan juga bahwa menggunakan teknik-teknik machine learning standar, seperti SVM dan Naive Bayes menghasilkan akurasi yang lebih baik dibandingkan dengan klasifikasi yang dilakukan oleh manusia. Namun lebih jelek dibandingkan kalau teknik-teknik itu digunakan untuk klasfikasi topik. Klasifikasi topik adalah "klasifikasi teks pada umumnya" atau lebih umum disebut kategorisasi teks. Dalam tulisan sentimen, ada hal-hal spesifik yang perlu digali/dioptimalkan agar akurasinya lebih baik, misalkan bagian-bagian kalimat yang menunjukkan opini. Sehingga salah satu pendekatannya adalah pada praprosesingnya (misalkan memanfaatkan NLP) dengan tetap menggunakan teknik machine learning standar. Sebagai contoh, diidentifikasi jenis kata yang dianggap mengindikasikan adanya opini.

bagaimana penelitian lebih lanjut tantang sentiment classification ini, al. bisa dilihat dengan Google Scholar dengan kata kuncil judul tulisan diatas "Thumbs up? Sentiment Classification using Machine Learning Techniques", kemudian klik tulisan apa saja yang merujuk ke tulisan ini. barusan saya lihat: ada 446 tulisan!

Terkait dengan sentiment classification adalah "opinion mining", yaitu menghasilkan summary dari tulisan, dimana summarynya adalah opinion yang ada di tulisan itu. Ini pekerjaan yang lebih sulit dibandingkan sentiment classification.

Friday, December 5, 2008

ICADIWT 2009, paper due Jan 10 2009

The Second International Conference on the Applications of Digital Information and Web Technologies (ICADIWT 2009)
London, United Kingdom, August 4-6, 2009

http://www.dirf.org/diwt2009/

IMPORTANT DATES

Submission Date: Jan 10, 2009
Notification of acceptance Two months from submission.
Camera-ready One month from notification.
Registration Two months from notification.
Conference date: August 4-6, 2009

AIPR-09, paper due: Feb 2 2009

AIPR-09
International Conference on Artificial Intelligence and Pattern Recognition

http://www.promoteresearch.org/2009/aipr/index.html


Draft paper submission due date is extended to: February 2 2009
Acceptance/rejection decision: March 2 2009
Camera ready paper submission and pre-registration: April 6 2009
Conference: July 13-16 2009

SAW 09, paper due: 1 Feb 09

Workshop on Social Aspects of the Web (SAW 2009)
http://bis.kie.ae.poznan.pl/12th_bis/wscfp.php?i=9&ws=saw2009

* Long papers: max. 12 pages
* Work-in-progress reports: max. 6 pages
* Demo papers: max. 4 pages

IMPORTANT DATES

* February 1, 2009 - submission deadline for papers
* February 22, 2009 - notification of acceptance/rejection
* March 15, 2009 - submission of final papers
* April 27, 28 or 29, 2009 - the workshop

Discovery Science 2009, paper due 10 may 09

http://ds09.liaad.up.pt/

The 12th International Conference on Discovery Science (DS-2009)

Important Dates
Submission deadline: 10 May 2009
Notifications: 21 June 2009
Camera-ready copy: 10 July 2009
Conference: 3-5 October 2009

Sunday, November 30, 2008

Educational Data Mining, paper due 31/3/09

International Conference on Educational Data Mining (EDM 09)http://www.educationaldatamining.org/EDM2009/

Paper submission: March 31, 2009
Acceptance notification: May 1, 2009
Camera ready paper: May 20, 2009
Conference: July 1-3, 2009

Saturday, November 29, 2008

AIEMPro09 London

Second International Workshop on Automated Information Extraction in Media Production (AIEMPro09) Special Session at WIAMIS 2009 (http://wiamis2009.qmul.net/)
London, 6-8 May 2009

Tentative deadlines:
Paper submission: 11 January 2009
Notification of reviews: 1 February 2009
Final camera ready (this is a STRICT DEADLINE): 13th February 2009

Areas of Interest (not limited to):
....
Automated news production
Content summarisation (e.g., sports highlights)
....

send the paperin PDF format DIRECTLY to the organisers by e-mail.
Organisers:
Alberto Messina (RAI CRIT) a.messina@rai.it
Jean-Pierre Evain (European Broadcasting Union) evain@ebu.ch
Robbie De Sutter (VRT medialab) robbie.desutter@vrt.be

Friday, November 28, 2008

IJCBS'09 Shanghai

The International Joint Conference on Bioinformatics, Systems Biology and Intelligent Computing (IJCBS'09) Shanghai, China on August 3rd - 6th, 2009

Important Dates
Paper Submission : February 15, 2009 (maximum 8 pages with IEEE/CS conference format)
Notification to Authors: March 15, 2009
Camera-Ready, Full Papers: April 5, 2009

mapping kata dasar..

Kalau ada corpus semacam PropBank, tampaknya juga bermanfaat kalau untuk kata dasar kita buat mapingnya: anatara kata dasar dengan turunannya, sehingga tidak perlu proses stemming. Dan membuat mapping ini tidak terlalu susah, hanya butuh waktu.

DEXA 2009

http://www.dexa.org

Database an Expert Systems Applications
August 31 – September 4, 2009

* Submission of abstracts: March 1, 2009
* Submission of full papers: March 8, 2009
* Notification of acceptance: May 22, 2009
* Camera-ready copies due: June 10, 2009

Wednesday, November 26, 2008

sentiment/opinion analysis

bacalah buku tentang opinion mining and sentiment analysis:
http://www.cs.cornell.edu/home/llee/opinion-mining-sentiment-analysis-survey.html

Tuesday, November 25, 2008

ide penelitian: menyusun "sejarah"(?)

Dari informasi2 internet (utamanya koran) kita susun secara semi otomatis suatu isi, biografi dll. Misal biografi Amien Rais. Termasuk dari koran yg diambil adalah opini.

MK studi terbimbing dan MK kompetisi

ide riset: Indonesian word list

http://indodic.com/index.html

Download our Indonesian word list for use in spell-checkers here. Read the Installation notes.

Bisa terus menerus ditambah dengan melakukan parsing web page yang menggunakan bahasa formal (misalkan koran). Lebih menarik kalau kemudian diklasifikasikan, misalkan apakah termasuk kata kerja, nama orang dst. Klasifikasi bisa hirarkis dan multilabel.

Lebih lanjut lagi juga mencakup yang non-formal (bahasa gaul) dan singkatan-singkatan.

Secara terus menerus mengambil kata yang ada dari beberapa situs web (misalkan dari koran-koran) diambil kata-kata yang dipakai. Kita buat daftar kata. Berapa frekuensinya. IDF. Kata-kata yang salah ketik.

Apa fungsinya, al:
pengganti "stemming"
membantu spelling checker
analisis, misal: kata yang sering dipakai, yang sering salah ketik...
xx
Kita bisa mencari keluarga kata itu.

menyederhanakan matrix pada cost sensitive learning

Banyak kasus riil (seperti pada Data Mining Cup 2008), nilai cost matrix pada TP, TN tidak bernilai nol. Sedangkan untuk beberapa penghitungan [jelaskan] dan juga beberapa keperluan praktis nilai2 cost untuk yang prediksinya benar mesti diisi nol (misalkan software SPSS Clementine mengharuskan seperti itu). Kita bisa mengubahnya dengan mengurangi dengan nilai yan sama untuk kolom yang sama.

[contoh, perlihatkan kasus dan cost matrix DMCup 2008)
[penjelasan penghitungan]
[referensi]

cost pada cost sensitive learning

Biasanya cost pada "cost sensitve learning" memang misclassification cost namum bukan hanya itu, .....

[penjelasan]
[contoh]
[referensi]