TY - GEN
T1 - FINCHAN
T2 - 2016 Annual Research Conference of the South African Institute of Computer Scientists and Information Technologists, SAICSIT 2016
AU - Ade-Ibijola, Abejide
N1 - Publisher Copyright:
©2016 ACM.
PY - 2016/9/26
Y1 - 2016/9/26
N2 - Instant messaging (IM) has gained huge popularity, not only in the social space, but in many sectors of work and every- day life. One such sector is finance, where IMs are used for bargaining, trading and many other financial information ex- changes between parties. A popular IM software in finance is the Instant Bloomberg (IB). The conversations (chat histo- ries) recorded via IB often contain crucial information about the financial positions of the market, quotes for di erent in- struments, as well as unimportant texts. Automatic compre- hension of this type of text is a natural language processing task. In this paper we employ front-end compiler analyses in the automatic comprehension and summarisation of financial chats retrieved from IB. Taking the result, we present a software tool called FINCHAN that takes lengthy chat his- tories and attempts to recognise the syntax up to a success rate of 98%, extracts semantic information to 61% success rate, generates concise summaries of these chats, and offers analyses of hidden patterns, text-to-speech synthesis of generated summaries, with functionalities for archiving/in dexing IMs.
AB - Instant messaging (IM) has gained huge popularity, not only in the social space, but in many sectors of work and every- day life. One such sector is finance, where IMs are used for bargaining, trading and many other financial information ex- changes between parties. A popular IM software in finance is the Instant Bloomberg (IB). The conversations (chat histo- ries) recorded via IB often contain crucial information about the financial positions of the market, quotes for di erent in- struments, as well as unimportant texts. Automatic compre- hension of this type of text is a natural language processing task. In this paper we employ front-end compiler analyses in the automatic comprehension and summarisation of financial chats retrieved from IB. Taking the result, we present a software tool called FINCHAN that takes lengthy chat his- tories and attempts to recognise the syntax up to a success rate of 98%, extracts semantic information to 61% success rate, generates concise summaries of these chats, and offers analyses of hidden patterns, text-to-speech synthesis of generated summaries, with functionalities for archiving/in dexing IMs.
KW - Comprehension
KW - Context-free grammar
KW - Financial instant messages
KW - Instant bloomberg
KW - Summarisation
UR - http://www.scopus.com/inward/record.url?scp=84994228063&partnerID=8YFLogxK
U2 - 10.1145/2987491.2987518
DO - 10.1145/2987491.2987518
M3 - Conference contribution
AN - SCOPUS:84994228063
T3 - ACM International Conference Proceeding Series
BT - SAICSIT 2016 - Proceedings of the 2016 Annual Research Conference of the South African Institute of Computer Scientists and Information Technologists
A2 - van der Haar, Dustin T
A2 - Coulter, Duncan A.
A2 - Coetzee, Marijke
A2 - Ehlers, Elize M.
A2 - Marnewick, Carl
A2 - Blauw, Frans F.
A2 - Leung, Wai Sze
PB - Association for Computing Machinery
Y2 - 26 September 2016 through 28 September 2016
ER -