Comparison between similarity measurments of vector space model on arabic text

Joint Authors

Kanan, Ghassan Jaddu
Hananidah, Isam

Source

Al-Manarah For Research and Studies

Issue

Vol. 16, Issue 3 (30 Jun. 2010), pp.73-85, 13 p.

Publisher

Al al-Bayt University Deanship of Academic Research

Publication Date

2010-06-30

Country of Publication

Jordan

No. of Pages

13

Main Subjects

Media and Communication

Topics

Abstract AR

في هذا البحث قمنا باختيار ملخصات من 242 وثيقة باللغة العربية و هذه الملخصـات كانت متخصصة في علمي الحاسوب و نظم المعلومات، لقد قمنا ببناء نظـام اسـترجاع لمعالجة البيانات باللغة العربية، و تم تطبيق تقنية الفهرسة التلقائية على الوثائق العربيـة، بحيث تم بناء النظام باستخدام نمـوذج المتجهـات الموجهـة de1m Space rVect باستخدام أربعة أصناف استخدمت حساب درجة التشابه بين الاستعلام و الوثيقـة و تمـت مقارنة النتائج لهذه الأصناف من المتجهات، و كانت النتيجة إن صنف الاحتمـال أظهـر أفضلية في عملية الاسترجاع مقارنة مع باقي أنواع المتجهات.

Abstract EN

This paper has selected 242 Arabic abstract documents which were used by (Hmeidi & Kanaan, 1997).

All these abstracts are about computer science and information systems.

We also designed and built an automatic information retrieval system from scratch to handle Arabic data.

The system was implemented in C# NET language, and Runs on IBM/PCs and compatible microcomputer.

An automatic indexing technique has been implemented for this corpus.

The system was built using Vector Space Model (VSM), In this model all mesurments were taken.

Cosine measure, Dice measure, Jaccard measure, and Inner product similarity were used.

The retrieval results using Vector Space Model were compared.

It was found out that the retrieval result is better than the retrieval result for in Arabic documents.

American Psychological Association (APA)

Hananidah, Isam& Kanan, Ghassan Jaddu. 2010. Comparison between similarity measurments of vector space model on arabic text. Al-Manarah For Research and Studies،Vol. 16, no. 3, pp.73-85.
https://search.emarefa.net/detail/BIM-327316

Modern Language Association (MLA)

Hananidah, Isam& Kanan, Ghassan Jaddu. Comparison between similarity measurments of vector space model on arabic text. Al-Manarah For Research and Studies Vol. 16, no. 3 (2010), pp.73-85.
https://search.emarefa.net/detail/BIM-327316

American Medical Association (AMA)

Hananidah, Isam& Kanan, Ghassan Jaddu. Comparison between similarity measurments of vector space model on arabic text. Al-Manarah For Research and Studies. 2010. Vol. 16, no. 3, pp.73-85.
https://search.emarefa.net/detail/BIM-327316

Data Type

Journal Articles

Language

English

Notes

Includes bibliographical references : p. 83-85

Record ID

BIM-327316