Conference Presentations by Gourab Nath
In the proceedings of Second International Conference on Frontiers of Operations Research & Business Studies (FORBS 2019), Calcutta Business School, Kolkata, West Bengal, India, 2019
As the market dynamics change, the market players also need to adapt or adopt the changes accordi... more As the market dynamics change, the market players also need to adapt or adopt the changes accordingly. That would require constant monitoring of marketing activities and updating positioning strategies. Market research is an important tool to understand how an organization is positioned in the market and if the positioning is improper, the organization might face difficulties in maintaining its revenue streams. Traditional market research is cost intensive from monitory point of view and also from the aspect of execution time. That is why consumer reviews can be an easy source of getting consumers view point and that can be used for understanding how organizations are positioned in the mind of the consumers. The current research tried to explore the possibilities of using consumer reviews for constructing perceptual maps. The focus was kept on perceptual map because it is an important tool in managing the brand.
Proceedings of 7th International Conference of Business Analytics and Intelligence, Indian Institute of Management, Bangalore, India, 2019
The purchase decisions of products sold online depend considerably on the reviews and ratings of ... more The purchase decisions of products sold online depend considerably on the reviews and ratings of the products given by the customers who have already purchased the product. This has been the common practice of the current time. Hence, the online reviews about the products have a strong influence on the purchase decisions of potential customers. Several false reviews may drive the influence of the customers either to a positive direction or to a negative direction. These may hamper the trustworthiness of the online opinions and the reliability of the e-commerce platform selling the products may decrease. Review spams, however, is not easy to detect. This problem is different from the email spam detection problem because spam emails are usually commercial advertisements and contain some very prominent features. In comparison, this problem is different as a spam review may be carefully crafted and made indistinguishable from the genuine reviews. One of the earliest attempts to detect spam reviews includes the use of supervised learning algorithm to detect spam by identifying duplicate reviews and extracting a set of effective data features of three different types (1) review centric features, (2) reviewer centric features, and (3) product-centric features Their research showed some tentative but interesting results. A lot of work has been done so far to address this problem. Shebuti and Leman [1] proposed a holistic approach called SpEagle [1] which uses clues from metadata (text, timestamp, rating) and relational data (network) to detect suspicious users and reviews, as well as products targeted by spam. In this paper, we aim to critically analyse a network-based approach to detect review spam and discuss our experimental results and the challenges faced.
Proceedings of 7th International Conference of Business Analytics and Intelligence, Indian Institute of Management, Bangalore, India, 2019
With the advancement over time the online shopping experiences are becoming remarkable because of... more With the advancement over time the online shopping experiences are becoming remarkable because of its convenience, better prices, easy price comparisons, no crowds and also because they are a big time saver. The customers in these platforms are highly encouraged to share their feedback about their experiences and the products they have purchased in the form of ratings and reviews. These reviews serve several purposes. First, this feedback may help potential customers to get a perspective of what they would like to purchase. Second, these help the manufacturer of the products to understand the customer satisfaction level. In this research, we intend to extend some of the existing works on sentiment analysis and suggest an effective way to cluster customer reviews for better and more informative summarization. Our task will help grouping similar reviews based on the intensity of the opinions expressed by the customers on various product features. Grouping similar review is similar to grouping similar customers based on their perspectives towards different product features. Therefore, we believe this work will help the manufacturers of the products to identify the customer satisfaction level belonging to different clusters and come up with solutions and/or strategies that fit the requirement of their customers to elevate the customer satisfaction level.
Proceedings of 7th International Conference of Business Analytics and Intelligence, Indian Institute of Management, Bangalore, India, 2019
At present WhatsApp is being used by 1.5 billion People. Though the platform is used to connect w... more At present WhatsApp is being used by 1.5 billion People. Though the platform is used to connect with family, friends and colleagues, it has been misused by a few to spread rumours and fake messages. When a user receives a WhatsApp forward message, he has very limited options at disposal to verify the veracity of the forward message. The user can attempt to verify with the forwarder or verify using google search. However, verifying the veracity of a WhatsApp message using google search is exhausting and the user cannot resort to google for all the forwards received. This sets the motif of this research. We will use Natural Language Processing to deeply explore the semantics of fake data and identify patterns in them using supervised Machine Learning techniques to distinguish between fake and genuine messages. We will also extend our exploration based on behavioral and contextual patterns of fake messages.
Proceedings of 7th International Conference of Business Analytics and Intelligence, Indian Institute of Management, Bangalore, India, 2019
"Buy Now-Pay Later; Interest-Free EMIs; Cashless Transactions". There is a whole new sanctum of e... more "Buy Now-Pay Later; Interest-Free EMIs; Cashless Transactions". There is a whole new sanctum of easy loans and credit wide open for every consumer due to the recent no-questions-asked policy. However, banks and credit firms all over the world are always under threat of being stood up when a borrower fails to repay a loan or defaults on their credit card bill. Many times, these loans are not backed by collateral and recovery is impossible. Since lending is the backbone of all banking businesses, a bad loan not only decreases the revenue and the profit of the bank but also adversely affects its reputation. Hence, all financial institutions need to do a Credit Risk Assessment of the applications for new loans and credit cards and discern the creditworthiness of the borrower. They use the credit history of the borrower and apply machine learning techniques to identify potential defaulters. The problem arises when this data turns out to be imbalanced. The objective of this project is to predict credit card default from a highly imbalanced dataset which includes only 4% default cases. The standard machine learning algorithms tend to get overwhelmed by the majority class and fail to identify the minority class and algorithm gets biased towards the non-default. Our primary challenge will be to improve the performance of selected algorithms by correcting the class imbalanced dataset to improve the model performances.
Papers by Gourab Nath
2021 International Conference on Intelligent Technologies (CONIT)
This research proposes a new recommender system algorithm for online grocery shopping. The algori... more This research proposes a new recommender system algorithm for online grocery shopping. The algorithm is based on the perspective that, since the grocery items are usually bought in bulk, a grocery recommender system should be capable of recommending the items in bulk. The algorithm figures out the possible dishes a user may cook based on the items added to the basket and recommends the ingredients accordingly. Our algorithm does not depend on the user ratings. Customers usually do not have the patience to rate the groceries they purchase. Therefore, algorithms that are not dependent on user ratings need to be designed. Instead of using a brute force search, this algorithm limits the search space to a set of only a few probably food categories. Each food category consists of several food subcategories. For example, "fried rice" and "biryani" are food subcategories that belong to the food category "rice". For each food category, items are ranked according to how well they can differentiate a food subcategory. To each food subcategory in the activated search space, this algorithm attaches a score. The score is calculated based on the rank of the items added to the basket. Once the score exceeds a threshold value, its corresponding subcategory gets activated. The algorithm then uses a basket-to-recipe similarity measure to identify the best recipe matches within the activated subcategories only. This reduces the search space to a great extent. We may argue that this algorithm is similar to the content-based recommender system in some sense, but it does not suffer from the limitations like limited content, overspecialization, or the new user problem.
Robust Portfolio Design and Stock Price Prediction Using an Optimized LSTM Model
2021 IEEE 18th India Council International Conference (INDICON), 2021
Accurate prediction of future prices of stocks is a difficult task to perform. Even more challeng... more Accurate prediction of future prices of stocks is a difficult task to perform. Even more challenging is to design an optimized portfolio with weights allocated to the stocks in a way that optimizes its return and the risk. This paper presents a systematic approach towards building two types of portfolios, optimum risk, and eigen, for four critical economic sectors of India. The prices of the stocks are extracted from the web from Jan 1, 2016, to Dec 31, 2020. Sector-wise portfolios are built based on their ten most significant stocks. An LSTM model is also designed for predicting future stock prices. Six months after the construction of the portfolios, i.e., on Jul l, 2021, the actual returns and the LSTM-predicted returns for the portfolios are computed. A comparison of the predicted and the actual returns indicate a high accuracy level of the LSTM model.
Prediction of future movement of stock prices has been a subject matter of many research work. On... more Prediction of future movement of stock prices has been a subject matter of many research work. On one hand, we have proponents of the Efficient Market Hypothesis who claim that stock prices cannot be predicted, on the other hand, there are propositions illustrating that, if appropriately modeled, stock prices can be predicted with a high level of accuracy. There is also a gamut of literature on technical analysis of stock prices where the objective is to identify patterns in stock price movements and profit from it. In this work, we propose a hybrid approach for stock price prediction using five deep learning-based regression models. We select the NIFTY 50 index values of the National Stock Exchange (NSE) of India, over a period of December 29, 2014 to July 31, 2020. Based on the NIFTY data during December 29, 2014 to December 28, 2018, we build two regression models using convolutional neural networks (CNNs), and three regression models using long-and-short-term memory (LSTM) netwo...
2021 International Conference on Intelligent Technologies (CONIT), 2021
This research proposes a new recommender system algorithm for online grocery shopping. The algori... more This research proposes a new recommender system algorithm for online grocery shopping. The algorithm is based on the perspective that, since the grocery items are usually bought in bulk, a grocery recommender system should be capable of recommending the items in bulk. The algorithm figures out the possible dishes a user may cook based on the items added to the basket and recommends the ingredients accordingly. Our algorithm does not depend on the user ratings. Customers usually do not have the patience to rate the groceries they purchase. Therefore, algorithms that are not dependent on user ratings need to be designed. Instead of using a brute force search, this algorithm limits the search space to a set of only a few probably food categories. Each food category consists of several food subcategories. For example, “fried rice” and “biryani” are food subcategories that belong to the food category “rice”. For each food category, items are ranked according to how well they can differen...
Uploads
Conference Presentations by Gourab Nath
Papers by Gourab Nath