Varun Tiwari

Varun Tiwari


Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications

31 January 2026 in Discover Computing
Verified
Journal Article
Open access
PDF file
CRChour Singh Rajpoot
Chour Singh Rajpoot
VTVarun Tiwari
Varun Tiwari
SVSANTOSH KUMAR VISHWAKARMA
SANTOSH KUMAR VISHWAKARMA
With the rapid growth of online information, the recommender system (RSs) has become essential tools for filtering information and generating personalized recommendations for customers. These systems analyze individual user preferences, past reviews, and product ratings to enhance decision-making in the e-commerce sector. In this comprehensive review, several research articles published between 2020 and 2025 have been reviewed from various journals and conferences, covering multiple dimensions of recommender systems. The study emphasizes emerging techniques aimed at improving system performance. It begins by providing background information on recommender systems, including existing literature and prevailing research challenges. Furthermore, it explores cutting-edge methodologies such as machine learning, deep learning, large language models, and conversational recommender systems that are contributing to the evolution of RSs. Beyond traditional approaches, this review addresses significant research gaps and identifies key aspects influencing RS platforms. The ultimate goal of this study is to present a comprehensive understanding of the current state and future direction of recommender systems in the e-commerce industry.

28 March 2026 in Discover Artificial Intelligence
Verified
Journal Article
Open access
PDF file
CRChour Singh Rajpoot
Chour Singh Rajpoot
VTVarun Tiwari
Varun Tiwari
SVSantosh Kumar Vishwakarma
Santosh Kumar Vishwakarma
E-commerce platforms generate vast amounts of data, making it challenging for customers to find relevant products efficiently. Recommendation Systems (RS) address this issue by analysing user interactions to suggest personalized items. The two primary RS techniques are Content-Based Filtering (CBF) and Collaborative Filtering (CF). However, CF struggles with tracking changes in user preferences and can perform poorly in sparse datasets. To overcome these limitations, this study proposes a hybrid recommender system that integrates both CBF and CF approaches to enhance recommendation accuracy. The system follows a structured process starting with data collection and preprocessing, which includes data cleaning, discretization, and Principal Component Analysis (PCA) to reduce dimensionality. Feature extraction then captures both content and collaborative features, utilizing methods such as profile construction, content similarity indexing (using Jaccard similarity), neighbor finding (with an improved Manhattan distance refined by Euclidean distance), and item weight generation. For optimizing recommendation ratings, the system employs a novel Adaptive Red Deer Optimization combined with Dandelion Optimizer (ARD-DO). Evaluation metrics such as Accuracy, Precision, Recall, MCC, F-measure, RMSE, MAPE, MSE, and MAE were used to assess performance. Experiments on two datasets show that ARD-DO outperforms other algorithms, achieving minimal MAPE (0.0134 and 0.027186) and RMSE (2.9914 and 2.9225), along with high accuracy (0.9757 and 0.9879). Comparative algorithms like RDO, DLO, BOA, and GFO showed inferior performance, confirming the effectiveness of the proposed hybrid system.

5 April 2024 in Journal of Machine and Computing
Verified
Journal Article
Open access
VAVenugopal Rao A
Venugopal Rao A
SVSantosh Kumar Vishwakarma
Santosh Kumar Vishwakarma
SKShakti Kundu
Shakti Kundu
VTVarun Tiwari
Varun Tiwari
Human activity recognition (HAR) is an active research area in computer vision from past several years and research is still continuing in this field due to the unavailability of perfect recognition system. The human activity recognition system it covers e-health, patient monitoring, assistive daily living activities, video surveillance, security and behaviour analysis, and sports analysis. Many researchers have suggested techniques that use visual perception to detect human activities. Researchers will need to address problems including light variations in human activity detection, interclass similarity between scenes, the surroundings and recording setting, and temporal variation in order to construct an efficient vision-based human activity recognition system. However, a significant drawback of many deep learning models is their inability to achieve satisfactory results in real-world scenarios due to the conflicts mentioned above. To address this challenge, we developed a hybrid HAR-CNN classifier aimed at enhancing the learning outcomes of Deep CNNs by combining two models: Improved CNN and VGG-19. Using the KTH dataset, we collected 6,000 images for training, validation, and testing of our proposed technique. Our research findings indicate that the Hybrid HAR-CNN model, which combines Improved CNN with VGG-19 Net, outperforms individual deep learning models such as Improved CNN and VGG-19 Net.


7 May 2025
Verified
Book Chapter
Closed access
PDF file
AGAaryan Gupta
Aaryan Gupta
PGPushkal Garg
Pushkal Garg
VTVarun Tiwari
Varun Tiwari
RPRanojit Palit
Ranojit Palit