Prof. P Krishna Reddy and his students presented the following papers at an International Conference on Database and Expert Systems Applications (DEXA-2021) held virtually from 27 – 30 September:
- Improving Billboard Advertising Revenue Using Transactional Modeling and Pattern Mining – Parvataneni Revanth Ratan, Krishna Reddy P and A Mondal
Research work as explained by the authors: Billboard advertisement is among the dominant modes of outdoor advertisements. The billboard operator has an opportunity to improve its revenue by satisfying the advertising demands of an increased number of clients by means of exploiting the user trajectory data. Hence, we introduce the problem of billboard advertisement allocation for improving the billboard operator revenue, and propose an efficient user trajectory-based transactional framework using coverage pattern mining. Our experiments validate the effectiveness of our framework.
- An Urgency-Aware and Revenue-Based Itemset Placement Framework for Retail Stores – Raghav Mittal (Ashoka University), Anirban Mondal (Ashoka University), Parul Chaudhary (Shiv Nadar University), Krishna Reddy P
Research work as explained by the authors: Placement of items on the shelf space of retail stores significantly impacts the revenue of the retailer. Given the prevalence and popularity of medium-to-large-size retail stores, several research efforts have been made towards facilitating item/itemset placement in retail stores for improving retailer revenue. However, they do not consider the issue of urgency of sale of individual items. Hence, they cannot efficiently index, retrieve and place high-revenue itemsets in retail store slots in an urgency-aware manner. Our key contributions are two-fold. First, we introduce the notion of urgency for retail itemset placement. Second, we propose the urgency-aware URI index for efficiently retrieving high-revenue and urgent itemsets of different sizes. We discuss the URIP itemset placement scheme, which exploits URI for improving retailer revenue. We also conduct a performance evaluation with two real datasets to demonstrate that URIP is indeed effective in improving retailer revenue w.r.t. existing schemes.
Link to the conference page: http://www.dexa.org/