WORKSHOP 2: MORPHOLOGICAL ANALYSIS OF PAKISTANI LANGUAGES


SPEAKERS

caret-down caret-up caret-left caret-right
Dr. Tafseer Ahmed Khan Muhammad Ali Jinnah University Karachi

Dr. Tafseer Ahmed got his PhD from Universitaet Konstanz, Germany in 2009. He has teaching and research experience of more than 17 years in various institutes including University of Karachi, FAST NUCES Lahore, University of Konstanz and DHA Suffa University. He is working on computational linguistics, text mining and machine learning. His goal is the development of solutions for Pakistani languages.

Dr. Mutee u Rahman Isra University, Hyderabad

Mutee U Rahman is Associate Professor in Computer Science at Isra University, Hyderabad. He is a member of Society for Natural Language Processing, Pakistan (SNLP). His research interests include Computational Linguistics and Natural Language Processing and processing of Pakistani languages. He holds a doctoral degree in Computational Linguistics from Isra University, Hyderabad and his Ph.D. thesis is on Computational Grammar Development of Sindhi Language in LFG framework.

INTRODUCTION

Morphology is the analysis of structure of the word and is an essential requirement for the determination of part of speech structure in syntax and semantic analysis of a sentence. Computational modeling of morphological analysis is an important resource for language processing applications. Some other applications of morphological analysis include: Internet search (information retrieval), Information Extraction, Machine Translation, and POS tagging. Foma and Python libraries are software tools available to model language morphology and perform morphological analysis. By using these tools useful linguistic resources for different languages can be developed. In this regard there is lot of work available on morphological analysis (including significant work on some Pakistani languages). Using this literature, we focus on creating common analysis and common standards (wherever possible) for Pakistani languages.  The workshop aims to present a brief introduction to the open source tools Foma and Python libraries related to morphology (including unsupervised morphological learning). This brief introduction will give awareness about the current state of the art tools for language morphology modeling, documentation and usage.

OBJECTIVES

(a) introducing current trends and technologies for morphological analysis
(b) creating/searching for a common method of analyzing Pakistani languages and
(c) networking among the practitioners and potential contributors.

WHO SHOULD ATTEND ?

Faculty Members and Scholars who are interested in new research topics of cutting edge language technology
Linguistics Students who want to know the basics of morphological analysis
Researchers who are interested in developing linguistic resources.

WORKSHOP PROGRAM

Time ActivityTime Facilitator
11:00 – 11:30Introduction: Morphology, Tagsets, and Features
Activity: Discovering morphological features
Dr. Tafseer Ahmed Khan
Dr. Mutee U Rahman
11:30 – 11:50Modeling Language MorphologyDr. Mutee U Rahman
11:50 – 12:00Demonstration on Foma
Introduction to PakMorph / Pakgram
Dr. Tafseer Ahmed Khan
Dr. Mutee U Rahman
12:00 – 12:30Python libraries related to MrophologyDr. Tafseer Ahmed Khan