Medical Language Processing: Computer Management of Narrative Data
Medical Language Processing: Computer Management of Narrative Data
Chronological information extraction system (chess)
Chronological information extraction system (chess)
MUC4 '92 Proceedings of the 4th conference on Message understanding
Mining Influential Bloggers: From General to Domain Specific
KES '09 Proceedings of the 13th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems: Part II
Utilizing social context for providing personalized services to mobile users
EuroSSC'10 Proceedings of the 5th European conference on Smart sensing and context
Mining influential bloggers: From general to domain specific, from explicit to implicit
International Journal of Knowledge-based and Intelligent Engineering Systems
| Hi-index | 0.00 |
This paper describes a metadata extraction technique based on natural language processing (NLP) which extracts personalized information from email communications between financial analysts and their clients. Personalized means connecting users with content in a personally meaningful way to create, grow, and retain online relationships. Personalization often results in the creation of user profiles that store individuals' preferences regarding goods or services offered by various e-commerce merchants. With the introduction of e-commerce, it has become more difficult to develop and maintain personalized information due to larger transaction volumes. is an NLP and Machine Learning (ML)-based automatic metadata extraction system designed to process textual data such as emails, discussion group postings, or chat group transcriptions. extracts both explicit and implicit metadata elements including proper names, numeric concepts, and topic/subject information. In addition, Speech Act Theory inspired metadata elements, which represent the message creators' intention, mood, and urgency are also extracted. In a typical dialogue between financial analysts and their clients, clients often discuss the items that they liked or have an interest. By extracting this information, constructs user profiles automatically. This system has been designed, implemented, and tested with real-world data. The overall accuracy and coverage of extracting explicit and implicit metadata is about 90%. In summary, the paper shows that an NLP-based metadata extraction system enables automatic user profiling with high effectiveness.