Identifying the Working Business Domain of a Region-Based Start-up using Localized Machine Learning Techniques
Arghya Kusum Das1, Susanta Mitra2
1Arghya Kusum Das, Department of Computer Science & Engineering, Techno International New Town, Kolkata, India.
2Dr. Susanta Mitra, Director, Amity University, Kolkata, India.
Manuscript received on 30 May 2019 | Revised Manuscript received on 07 June 2019 | Manuscript published on 30 June 2019 | PP: 1933-1937 | Volume-8 Issue-8, June 2019 | Retrieval Number: G5599058719/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Initiating a startup totally revolves around the domain/area of work the start-up is expected to address. It is easy to start a startup provided the necessary infrastructure is available, but maintaining the startup in the long run is difficult as it depends on the successful operability of the startup. Hence to initiate the start-up the domain fixation is very important which is dependent on the geo-socio-economic factors. In this document, before starting the operations of a start-up, a survey is done by following the tweets/posts related to the locality specifically done to capture the lacunas of the region based on which the working domain of the startup would be finalized. For every input which is in the form of tweets/posts/blogs extracted from the different online platforms in social media the input text is parsed into tokens and finally the sentence is classified into an appropriate pre-defined level. If the classification level is found to be negative, then the appropriate agent is analyzed and finally the proposed requirement is found. The proposed work first identifies the negative tweets/posts. On each such negative tweets/posts thus extracted, the algorithm tries to find the lacunas and finally proposes the requirements. There may be multiple requirements for a specific region/location. In such scenarios, the algorithm clusters the similar requirements together and finally the largest cluster thus formed is concluded to be the most desired requirement which can later be identified as the working domain of the start-up which is fixed and finalized depending on the geo-socio-economic factors
Keywords: Classification, Clustering, Polarity, Sentiment Analysis, Localized Text Processing
Scope of the Article: Classification