Research Interests: Sentiment Analysis Applications, Social Media Analytics, Panel Data Modelling, ICT diffusion in SMEs, and Multi-Criteria Decision Making
Amit Singh holds a PhD degree from the Department of Industrial and Systems Engineering, Indian Institute of Technology Kharagpur (IIT Kharagpur), India. He received his BTech from the Department of Mechanical Engineering, National Institute of Technology Agartala (NIT Agartala). He obtained his MTech in Industrial and Management Engineering from the National Institute of Technology Jamshedpur (NIT Jamshedpur), India and was awarded a bronze medal. He has served organizations such as the Indian Institute of Technology Jodhpur, Quantium Analytics, and IBS Hyderabad before joining Ahmedabad University. His research interests include sentiment analysis applications, social media analytics, panel data modelling, ICT diffusion in small and medium enterprises, multi-criteria decision-making, and supply chain management, among others. He has published several research articles in leading international journals such as the Journal of Business Research, the Journal of Enterprise Information Management, the Journal of Global Information Management, the Journal of Modelling in Management, etc. He is a reviewer for leading journals such as the International Journal of Information Management, the Government Information Quarterly, the Information Systems Frontier, the Journal of Modelling in Management, the International Journal of Electronic Government Research, etc.
Algorithms to extract features/topics from reviews and quantify corresponding sentiments.
This algorithm extracts attribute/feature-level consumer sentiment from online car reviews. It comprises two steps. First, it uses a semi-automatic approach for feature extraction. This step cleans and processes reviews using the Rapid Automatic Keywords Extraction (RAKE) algorithm, followed by manual scanning to shortlist exact car attributes/features. Next, it targets the shortlisted attributes, searches corresponding sentiment-bearing words, and creates a list of subsentences consisting of only one attribute and corresponding opinion-bearing word(s). Finally, using the SentiWordNet dictionary, sentiments in the subsentences are quantified and assigned to the attribute therein.
A text analytics framework for consumers' perceived performance assessment of manufacturers.
This framework uses the sentiments extracted using the algorithm reported in the previous section. Here, we have clubbed the sentences of similar words under the broader umbrella named aspect, and the aspect level positive sentiment index was computed for each manufacturer for each quarter. Next, we integrate the sentiments concerning various manufacturers and aspects with the Technique for Order Preferences by Similarity to the Ideal Solution (TOPSIS), and House of Quality (HoQ) of Quality Function Deployment (QFD) to prioritize the manufacturers based on their consumer perceptions.
Integrated text analytics framework for vehicle weakness detection
Here we have proposed two different frameworks. The first framework feeds the sentiments extracted in Section A to the control charts, identifies the perceived weaknesses at aspect and attribute levels, and visualizes them using fishbone diagram. The second one uses Pareto charts to discover consumers' perceived weaknesses.
Understanding the influence of consumer-perceived negativity on car sales.
In this framework, we used aspect-level sentiments, and car sales data for the mid-sized car segment in India. We analysed them using a bias-corrected least square dummy variable estimator of panel data regression to understand if consumer perceptions expressed in car reviews influence sales. We have also quantified consumer sentiments concerning various supply chain partners for different manufacturers and studied if consumer perception propagates over time and influences sales.
Understanding the interdependencies among hindrances in the technology adoption
Here, we have developed three frameworks to analyze the interdependencies among the technology adoption barriers. The first framework analyses ICT adoption barriers in Indian SMEs; the second one analyses the adoption of lean manufacturing in SMEs, and the third one analyses integrated RFID-Blockchain adoption in public distribution systems.