HBHimani Bansal

Himani Bansal

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Himani Bansal has a total of 23 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Resource-efficient federated machine unlearning via evolutionary synaptic pruning for cloud-based distributed learning systems

    by Himani Bansal, Bhuvan Unhelkar, Dilip Kumar Jang Bahadur Saini, Bipin Kumar Rai, Pooja Sharma, Gautam Kumar, Prasun Chakrabarti

    With an increasing emphasis on user data control and privacy regulations, such as the General Data Protection Regulation (GDPR), machine unlearning (MU) has emerged as a crucial mechanism for managing data in AI systems. MU enables models to remove the influence of specific user data upon request. This problem becomes more complex in collaborative settings such as Federated Learning (FL), where data remains distributed across multiple clients, giving rise to Federated Unlearning (FU). In large-scale deployments, particularly those supported by cloud infrastructure, retraining models to satisfy data deletion requests can be computationally expensive, energy-intensive, and disruptive to ongoing services. This highlights the importance of having effective methods to unlearn outdated practices that can hinder the growth of a system and compromise data privacy awareness. We propose PRUNE-FL (Privacy-preserving Retention-focused Unlearning with Neuro-Evolution in Federated Learning), a framework that uses relevance-guided pruning and evolutionary optimization to delete the influence of the targeted data. PRUNE-FL is different from methods that rely on retraining or coarse parameter updates because it focuses on finding and changing the parameters that are most closely related to the data that needs to be forgotten. The approach is based on a synaptic relevance scoring system to figure out how each model parameter relates to certain client or class-level data. This makes it easier to find parameters that are connected to the target data. Then, the unlearning task is set up as a multi-objective problem to find a balance between the overall performance and the forgetting of unnecessary information. Finally, a genetic algorithm is used to implement an evolutionary pruning strategy. It seeks the optimal pruning settings that operate within the constraints of federated learning. Thus, PRUNE-FL helps in unlearning specific data without having to retrain the whole system. Tests on the CIFAR-10 dataset in both IID (Independent and Identically Distributed) and non-IID settings show that PRUNE-FL has higher accuracy. It also effectively removes the influence of the targeted data. The results also show it to be strong against bad patterns like backdoor triggers. Overall, PRUNE-FL enhances privacy by selectively unlearning the data and using fewer resources in federated environments.

  • From Forgetting to Future: A Survey of Machine Unlearning Approaches

    by Sheetal Sehgal, Ankita Verma, Himani Bansal

    The Right to Erasure has facilitated the erasure of data as part of ethical and legal compliance for all users. In this regard, Machine Unlearning (MUL) is an emerging tool that transforms the existing trained model to execute comprehensive data erasure. Although beneficial to a few, the rapid development of unlearning algorithms has hindered a beginner's ability to recognize the relationship between the algorithm's productivity and model effectiveness. Though machine unlearning can be quite instrumental in building ethical and trustworthy AI, it is not beyond corrections. Substantive gaps in the form of standardization, formal guarantees, and verifiable implementation do exist. The aim of this paper, therefore, is to present a comprehensive understanding of the machine unlearning field literature and practices. Over 100 peer‐reviewed articles were reviewed, available from 2019 to 2025, including those on federated learning, continual learning, graphical neural networks, and rapidly growing models like generative and large language models. The paper presents a critical analysis of evaluation metrics, unlearning verification, efficiency, model utility on retained data, scalability, and handling of interdependent and multimodal data. Furthermore, it proposes future research directions, pressing the need for effective and efficient unlearning mechanisms. This article is categorized under: Technologies > Machine Learning


  • Open Access

    Tampering with Truth: Designing a Tool to Undermine Digital Evidence Recovery

    by Himani Bansal, Manju Nunia

    The internet and technology advancements have given rise to too much comfort for both legitimate and malicious users or cyber criminals. Therefore, if these advancements are misused deliberately by the offenders, it can result in many harmful consequences, such as the prevention of services for benign or legitimate users. It is very difficult to investigate and prosecute any electronic crime because investigators need to build their cases or perform the investigation according to the evidence left by the computer criminals on the system. These days, computer criminals or adversaries are very much aware of the Computer Anti-forensic tactics and methods. Criminals apply such anti-forensic techniques to impede the whole digital investigation process efficiently and successfully. Such forensic investigation processes affected by Anti-forensic measures are too expensive and time-consuming to carry out. Numerous anti-forensic techniques can be used by anti-forensic practitioners to thwart the investigation process. Therefore, intruders try to hide or wipe out the evidence from the compromised system so that they cannot fall into the hands of Forensic examiners. Adversaries can try their very best to protect their evidence so that they can use any anti-forensic techniques for the same. In this study, we have proposed an Anti-Forensic tool for destructing the evidential files stored on the hard disk. Our proposed tool works under the controlled environment of the Secondary extended file system of the Linux distribution. This tool clears out all the inodes which actually store the metadata of the files and folders on the file system, so clearing the inodes corrupts the file system structure in the internal structure of the file system. Detection of such activity performed by the anti-forensic practitioner is possible by the Forensics software or Forensic Investigator during the Investigation. Still, it will not be possible for the analyst to recover the file content as the file will no longer be accessible, as its inode entries have already been cleared out by the intruder. In this research work, we have also compared our proposed tool with similar existing tools and found that none other than the proposed one could clear out all the inode entries of the evidential files on the File System.

  • Retraction
    Open Access

    Retraction Note: Crowd analytics: literature and technological assessment

    by Himani Bansal, Kavita Sharma, Manju Khari

  • Open Access

    MEDIBOT-An AI Powered Chatbot

    by Smriti Sharma, Nishita Rai, Saumya Bansal, Himani Bansal

    In today's fast-paced healthcare landscape, the demand for accessible, efficient, and personalized medical assistance is rapidly increasing. However, healthcare professionals often face challenges providing round-the-clock patient support during busy schedules. This leads to difficulties in promptly addressing inquiries and disseminating accurate information and may encounter obstacles in accessing timely medical advice, especially during non-office hours or in remote locations. To bridge this gap and enhance healthcare accessibility, developing a Language Model (LLM) trained chatbot tailored specifically for healthcare purposes is essential. This chatbot is designed to provide users with a reliable platform for seeking medical advice, information about symptoms, medication queries, and general healthcare guidance. By leveraging advanced natural language processing capabilities, the chatbot aims to address the needs of both healthcare professionals and patients alike.


  • Retracted

    RETRACTED ARTICLE: Crowd analytics: literature and technological assessment

    by Himani Bansal, Kavita Sharma, Manju Khari


  • Trust and Credibility Analysis of Websites

    by Himani Bansal, Prakhar Shukla, Manav Dhar

    Trust on any online information is psychosomatic and hidden by nature. The choice is in the hands of the information seeker to consider, evaluate, and confirm the contents of the websites before using it. This makes a sharp concern for websites dealing with sensitive topics like health, research, or academics. There is no benchmark or tool that tells or characterises about making these “trust” decisions. Although web users make such decisions after considering numerous factors, still there are no such criteria to fulfil the underlying principle to deal with such decision making. This chapter is an effort to resolve the problem of how to measure the content provided by any website in terms of its credibility. Various models have been projected in this chapter to identify several factors pertaining to the credibility of content and users' trust on any website and accordingly analyse the identified factors to assess the websites.