Theme Description

Theme of the Symposium
In Natural language Processing, Parsing is considered to be a very important component that involves a number of issues concerning grammatical formalisms, algorithmic complexities, ambiguity resolution, and so on. There are many approaches adopted by the Parsing community, such as stochastic parsing or classical deep/full parsing. The task is at hand range from segmentation into chunks to deep parsing based on wide coverage grammars, including shallow and/or robust parsing.
In deep/full parsing, sentences are assigned a complete syntactic structure. However, not all natural language applications demand a complete syntactic analysis. A deep parse often provides more information than actual requirement, and sometimes less. Shallow/partial parsing, on the other hand, is a task of recovering only a limited amount of syntactic information from natural language sentences. Key parts of the syntactic structure or key pieces of semantic information are identified or extracted in shallow parsing, instead of producing a detailed syntactic or semantic analysis of each sentence. In particular, such tasks include identifying the noun phrases in a text and extracting non-overlapping chunks, the subject, the main verb and the object from a sentence. Typical modules in a shallow parser include Parts-of-speech tagging (given a word and its context, the correct morphosyntactic class of that word is determined), chunking (given the words and their morphosyntactic class, a chunk is determined) and morphosyntactic tagging (given the chunks of a sentence, the syntactic relationships among the chunks are determined).
Shallow parsers are used to reduce the search space for full blown deep parsers. These have been proved to be useful in the application domains like Information Retrieval, Summarization, Question-Answering, etc. Other applications of shallow parsing include data mining from unstructured textual material from the web, automated annotation of linguistic corpora and the preprocessing of data for high level linguistic tasks.
Like any other NLP tasks, shallow/partial parsing of Indian languages is also a challenge. Over the last decade there has been an increased interest in Parts-of-speech tagging and chunking of Indian languages. The stress is now put on shallow parsing, which, in turn, will be a push for deep/full parsing of Indian languages.
Original and unpublished research papers are solicited from academia and industry which address linguistic, foundational, or computational issues relating to Shallow Parsing in the context of Indian Languages Processing, including but not limited to:
morphology, syntax, morphosyntax, semantics, pragmatics
linguistic and mathematical models for Indian languages
text segmentation and preprocessing
morphological processing
parts-of-speech tagging
morphosyntactic tagging
designing principles for shallow-parsed large corpora
chunking and partial parsing of large amounts of texts in Indian Languages
corpus-based language modeling
language-oriented machine learning
software systems for management and accessibility to shallow-parsed large      
applications of shallow-parsed large Indian language corpora.
interpreting and generating written Indian languages
multi-lingual processing and machine translation
shallow processing in NLP Applications
Indian Language Speech Processing using NLP Capabilities

Centre for Indian Language Technology (CFILT)
Department of Computer Science and Engineering
Indian Institute of Technology Bombay
Powai, Mumbai-400076, India
Phone: +91-22-2576 4729 / 7718
Fax: +91-22-25720290 /25723480

-----------------------------------------------last updated 21 FEB 2006------------------------------------------------------------