Invited Talks
Topic |
|
|
Internet Advertising Using Sentiment Analysis |
|
|
|
|
Date |
|
|
1th May, 2014
|
|
|
|
|
Venue |
|
|
|
Topic |
|
|
Sentimental Intelligence |
|
|
|
|
Date |
|
|
29th Feb, 2013
|
|
|
|
|
Venue |
|
|
|
Topic |
|
|
Sentiment Analysis-The Hard Problem |
|
|
|
|
Date |
|
|
21th November, 2012
|
|
|
|
|
Venue |
|
|
|
Topic |
|
|
Commercial Possibilities of Sentiment Analysis |
|
|
|
|
Date |
|
|
30th May, 2011
|
|
|
|
|
Venue |
|
|
|
Topic |
|
|
Sentiment…Human Intelligence |
|
|
|
|
Date |
|
|
3rd February, 2012
|
|
|
|
|
|
Venue |
|
|
|
|
|
|
|
Abstract |
|
Sentiment Analysis (SA) from natural language text is a multifaceted and multidisciplinary AI problem, tries to narrow the communicative gap between the highly sentimental human and the sentimentally challenged computer by developing computational systems that recognize and respond to the sentimental states of the human users. There is a perpetual debate about better way of collecting intelligence either by following the functional path of biological human intelligence or generating new methodologies for completely heterogeneous mechatronics machine and redefine a completely new horizon called electronic intelligence. All the research endeavors till date try to find out the optimum solution strategies for machine that either mimic the techniques of self-organized biological human intelligence or can at least simulate the functional similarities of human sentimental intelligence.
SA defines an overall problem, which address multiple aspects of sub-problems. Human sentiment knowledge grows with its age and daily cognitive interactions. Therefore an intelligent should need some prior knowledge to act properly. Sentiment knowledge acquisition generally wrapped into computational lexicon, technically called Sentiment Lexicon. As like classical pattern recognition problem SA is also classified into identification and classification genre called subjectivity detection and polarity classification involves sentiment detection and sentiment classification. Proper structurization is required to proceed for any further granular analysis. Structurization involves identification of sentiment holder, sentiment topic and so on. The philosophical notion of engineering is always: “Necessity is the mother of all invention”. Sentiment/opinion aggregation is necessary requirement at the end user’s point. For example, an end user might desire an at-a-glance presentation of the main points made in a single review or how opinion changes time to time over multiple documents. To do so there are multiple aspects of the research needs more attention. On real-life applications, to provide a completely automated solution is the ultimate desired goal of all the sentiment analysis research. An intelligent system should smart enough to aggregate all the scattered sentimental information from the various blogs, news article and from written reviews. The role of any automatic system is to minimize human user’s effort and produce a good sensible output. Therefore textual or visual summarization or tracking of sentiment is the striking need at the end user’s point.
|
Topic |
|
|
Let’s Create Sentiment Resources for my Own Language |
|
|
|
|
Date |
|
|
30th May, 2011
|
|
|
|
|
Venue |
|
|
|
Abstract |
|
|
Sentiment analysis (SA) from natural language text is a multifaceted and multidisciplinary problem simultaneously. SA defines an overall problem, which address multiple aspects of sub-problems. Human sentiment knowledge grows with its age and daily cognitive interactions. Therefore an intelligent system should need some prior knowledge to act properly. Sentiment knowledge acquisition generally wrapped into computational lexicon, technically called Sentiment Lexicon. The other necessity for the SA task is human annotated corpus.
As a member of NLP community we all are well familiar with the natural language data insufficiency. Any research or development issue for NLP systems demand more and more annotated or un-annotated data for training, testing, evaluation or enrichment of the system. Indian NLP community is not an exception rather it is a challenging issue to collect multilingual data. India is the highest variant language speaking country, where the numbers of official languages are presently 18. Most of the Indian languages are electronically resource scarced, therefore resource creation is the striking need to start Sentiment Analysis research for any of the Indian languages.
|
Topic |
|
|
Sentiment Analysis |
|
|
|
|
Date |
|
|
3rd March, 2011
|
|
|
|
|
Venue |
|
|
|
|
|
|
|
Abstract |
|
|
Sentiment analysis (SA) from natural language text is a multifaceted and multidisciplinary problem simultaneously. SA defines an overall problem, which address multiple aspects of sub-problems. Human sentiment knowledge grows with its age and daily cognitive interactions. Therefore an intelligent system should need some prior knowledge to act properly. Sentiment knowledge acquisition generally wrapped into computational lexicon, technically called Sentiment Lexicon. As like classical pattern recognition problem SA is also classified into identification and classification genre called subjectivity detection and polarity classification involves sentiment detection and sentiment classification. Proper structurization is required to proceed for any further granular analysis. Structurization involves identification of sentiment holder, sentiment topic and so on. The philosophical notion of science is always: “Necessity is the mother of all invention”. As mentioned in the first paragraph that information processing need drive us to develop such system, meet user satisfaction level. Therefore textual or visual summarization or tracking of sentiment is the striking need at the end user’s end.
|
Presentations in Conferences
Sentimantics: The Conceptual Spaces for Lexical Sentiment Polarity Representation with Contextuality (Oral)
In The 3rd Workshop on Computational Approaches to Subjectivity and Sentiment Analysis (WASSA), ACL 2012, Jeju, Korea
The 5W Structure for Sentiment Summarization-Visualization-Tracking (Poster)
In The 13th International Conference on Intelligent Text Processing and Computational Linguistics (CICLING 2012), Delhi, India
Dr Sentiment Creates SentiWordNet(s) for Indian Languages Involving Internet Population (Oral)
In The 3rd IndoWordNet Workshop, ICON 2010, Kharagpur, India
Can We Mimic Human Pragmatics Knowledge into Computational Lexicon? (Oral)
In The 8th International Conference on Natural Language Processing (ICON 2010), IIT Kharagpur, India
Semantic Role Labeling for Bengali Noun using 5Ws (Oral)
In The 6th IEEE International Conference on Natural Language Processing and Knowledge Engineering (IEEE NLP-KE'10), Beijing, China
SentiWordNet for Indian Languages (Oral)
SemanticNet-Perception of Human Pragmatics (Oral)
Clause Identification and Classification in Bengali (Oral)
Topic-Based Bengali Opinion Summarization (Poster)
In The 23rd International Conference on Computational Linguistics (COLING 2010), Beijing, China
Detection in English and Bengali: A CRF‐based Approach (Poster)
In The 7th International Conference on Natural Language Processing (ICON-2009), Hyderabad, India
English to Hindi Machine Transliteration at NEWS 2009 (Poster)
In The ACL_IJCNLP-2009, Singapore
English Bengali Ad-hoc Monolingual Information Retrieval (Oral)
In The Forum for Information Retrieval Evaluation (FIRE-08), Kolkata, India
Language Independent Named Entity Recognition in Indian Languages (Oral)
In The 3rd International Joint Conference on Natural Language Processing (IJCNLP-08), Hyderabad, India