Dr Vimal Babu, Associate Professor in the Department of Management along with his PhD scholar Ms Rukma R, have co-authored a research paper titled “Enhancement of New Random Forest Algorithm to Predict the Employee Attrition Rate.” Published in the International Journal of Enterprise Network Management (ABDC-B), the paper also includes contributions from Dr Vijaya Prabhagar M, Assistant Professor at IIM Jammu.

Their work focuses on improving machine learning techniques to better predict employee attrition, that is, the employees who are likely to leave the organisation. The study focuses on the impact of HR analytics adoption by applying algorithms to predict employee attrition. It uses the Random Forest algorithm to identify reasons for attrition. However, this algorithm can be slow due to the large number of decision trees it uses. To improve it, the study proposes a new method that selects the best decision trees using a technique called fractional factorial design, which outperforms all others in predicting employee attrition.

Abstract

The problem of employee attrition in every organisation concerns the employee turnover ratio, thereby increasing the cost of investment in human resources. Various factors are reasonable for the rapid attritions at different phases. The purpose of the current study is to predict the employees who are likely to leave the organization. The factors that lead to attrition are identified using the Random Forest algorithm. The Random Forest algorithm is a widely used supervised machine-learning technique for classification and prediction. However, the random forest algorithm has certain issue like it is too slow and ineffective for real time predictions. i.e., the large number of trees can make the algorithm, which results in a slower model. Therefore, the study proposes, a new alternative for choosing the appropriate decision trees based on the concept of fractional factorial design of experiments. The different performance criteria were compared across the modified random forest algorithm, existing random forest algorithm, Support Vector Machine (SVM), Logistic Regression (LR), Navie Bayes (NB), K – Nearest Neighbour (K-NN), Decision Tree, XG Boost tree and Neural Network (NN). It was found that the modified random forest algorithm excelled in all criteria and performed better than the existing ones.

Practical Implementation/Social Implications of the Research

Practical implication: It emphasises the importance of focusing on factors like Business Travel, Job Role, Over Time, Department, Marital Status, and Education to reduce employee attrition. Organisations should design retention programs that support frequent travellers, manage overtime effectively, and address specific departmental needs. Tailoring job roles, promoting work-life balance through flexible hours, and providing educational opportunities are key strategies for enhancing employee satisfaction and retention.

Social Implication: The measures of this study can transform work culture into a more supportive and inclusive environment, promote gender equality by supporting employees with family responsibilities, and contribute to economic stability through reduced attrition. By investing in employee well-being, organizations can also strengthen their community engagement and corporate social responsibility initiatives for both employees and the broader society.

Future Research Plans

Future research could explore other factors influencing employee attrition, such as organizational culture, leadership style, and employee engagement. Longitudinal studies could track changes over time to understand the long-term effects on attrition rates. Additionally, investigating the interplay between identified factors, such as how Job Role and Over Time influence attrition, could provide deeper insights. Expanding the study across different industries and locations may reveal specific trends, allowing for tailored retention strategies. Lastly, incorporating qualitative methods like interviews and focus groups could offer a richer understanding of employees’ experiences and perceptions, complementing the quantitative findings.

With the increasing emphasis on startup culture and enhancing entrepreneurial ventures among the youth, Dr Aftab Alam, Assistant Professor from the Department of Management has published a paper “Developing a Reflective-formative-formative Scale for Measuring University Entrepreneurial Ecosystem from Students’ Viewpoints” in the Q1 journal IEEE Transactions on Engineering Management having an impact factor of 4.6.

Dr Aftab and his team have conducted research on how universities can create the best environment to support entrepreneurship among students and staff. The team has identified three key elements that help an entrepreneurial ecosystem thrive:

1. Skill Development – Helping people learn how to start and manage businesses.
2. Resources – Providing tools, funding, and guidance.
3. Culture – Encouraging creativity, risk-taking, and innovation.

They have also created a way to measure how well universities are doing in building this supportive environment, helping them improve and compare with others.

Abstract

This research addresses the gap in the literature on University Entrepreneurial Ecosystems (UEE) by conceptualizing UEE as an ecology-inspired system with dimensions like Entrepreneurial Skill Development, Resources, and Culture. Unlike prior studies focusing on isolated aspects, we provide a comprehensive framework and develop robust measurement scales using a rigorous four-step methodology, ensuring nomological validity. The paper contributes to entrepreneurial ecosystem literature by offering a novel conceptualisation and practical tool for evaluating and comparing UEE performance. This study aids scholars in understanding UEE’s holistic impact and supports managers in enhancing university-driven entrepreneurship for regional development.

Practical Implication of the Research

This paper contributes to the entrepreneurial ecosystem literature by conceptualising and measuring performance of university entrepreneurial ecosystems (UEE). Beyond academic circles, it could also serve as a valuable tool for managers seeking to evaluate and enhance the performance of these ecosystems.

Collaboration

1. Dr Arpita Ghatak, Kent Business School, University of Kent
2. Dr Bhaskar Bhowmick, Indian Institute of Technology Kharagpur
3. Dr Swagato Chatterjee, Queen Mary University of London

Future Research Plan

1. Longitudinal Validation: Conduct longitudinal studies to evaluate how UEE dimensions evolve over time and their sustained impact on entrepreneurial outcomes like spinoffs, startups, and regional development.
2. Comparative Analysis: Compare UEEs across diverse geographic, cultural, and institutional contexts to identify patterns, best practices, and challenges unique to different ecosystems.
3. Entrepreneurial Outcomes Measurement: Develop advanced metrics to assess the direct and indirect outcomes of UEEs on entrepreneurial intentions, skill-building, and economic growth.

Read more

Confirmatory first-order measurement model results for the subscales

Hierarchical Third-Order Measurement Model Results

The patent titled “System and Method for Controlling the Dissemination of Data on Digital Platforms,” authored by Assistant Professor, Dr M Asadul Haque from the Department of Management and his research Scholar Mr Suman Kumar Tiwari offers an innovative approach to enhancing advertising strategies on digital platforms. With application number 202541004114, their patent outlines a comprehensive system that utilises real-time data collection and analysis to optimise consumer engagement and improve brand effectiveness in the ever-evolving landscape of social media advertising.

Abstract

The patent outlines a novel system and method for optimizing the dissemination of data on digital platforms, specifically targeting the enhancement of advertising strategies through real time data collection and analysis. This system leverages APIs and web scraping for continuous, up-to-date consumer behaviour monitoring, enabling brands to make immediate adjustments to their advertising tactics based on current trends and insights. It employs advanced machine learning algorithms and sentiment analysis to detect intricate consumer response patterns, offering a nuanced understanding of advertisement effectiveness. Additionally, the system supports dynamic strategy optimization, facilitating real-time adjustments to advertising efforts in response to market and consumer behaviour shifts. It incorporates a holistic evaluation approach, considering various mediating and moderating factors such as consumer attitudes and platform types. Designed for scalability across multiple social media platforms, the system ensures compliance with data privacy regulations while performing real-time data collection and analysis. Ultimately, this patent presents a comprehensive feedback system aimed at improving consumer engagement and conversion rates through personalized and effective advertising strategies.

Explanation in Layperson’s terms

The research presents an innovative system aimed at improving advertising strategies on social media platforms. This smart system uses data analysis and machine learning to help brands understand the impact of their ads on consumer buying decisions. By collecting and analysing data from social media, it considers various factors such as ad content, social trends, and consumer sentiments. The system can predict the effectiveness of different advertising strategies before they are implemented, enabling brands to make informed decisions.

Traditional methods focus on basic metrics like ad clicks, but this invention provides a comprehensive view of what influences consumer behavior. It allows brands to adjust their advertising strategies in real-time based on consumer reactions, making campaigns more effective.

Benefits for businesses include better consumer engagement, higher sales, data-driven decision-making, and more efficient use of advertising budgets. The technology is particularly valuable for marketing agencies, online retailers, consumer product companies, and tech startups focused on marketing solutions.

Practical Implementation

The practical implementation of the new advertising system involves several key steps to ensure its effectiveness and ethical use. Firstly, the system integrates with existing social media platforms and advertising tools like Facebook Ads and Google Ads, enhancing advertising strategies. To ensure successful adoption, companies need to provide comprehensive training and support for their marketing teams, enabling them to use the new tools and interpret insights effectively. Before a full-scale rollout, businesses can conduct pilot programs to test the system on a smaller scale and make necessary adjustments based on real-world results. Ensuring adherence to data privacy regulations, such as GDPR, is crucial for the responsible and ethical handling of consumer data. Additionally, the system can be customized to meet the specific needs of various industries, by analysing sector-specific consumer behaviour and trends, which ensures tailored and effective advertising strategies.

Social Implications

The social implications of implementing the new advertising system are multifaceted. Firstly, it enhances consumer awareness by enabling brands to create more relevant and personalized advertisements, thus improving consumer experience and engagement. However, there is a potential risk of market saturation, where the overuse of similar strategies by multiple brands could overwhelm consumers with excessive ads. Privacy concerns are also significant, as increased data collection necessitates a careful balance between effective targeting and respecting consumer privacy. Additionally, automation and advanced analytics may shift job roles within marketing teams, prompting the need for reskilling towards more strategic, creative, and analytical tasks. The system promotes ethical marketing practices by leveraging data-driven insights to prevent misleading advertisements and ensure truthful marketing messages. Lastly, by understanding consumer sentiment, brands can influence societal trends and public perceptions, particularly through socially responsible marketing efforts.

Future Research Plans

• Exploration of Data Privacy Solutions- Investigate methods to enhance data privacy and security while still leveraging consumer data for effective advertising.

• Longitudinal Studies on Consumer Behaviour- Conduct longitudinal studies to understand how consumer behaviour changes over time in response to targeted advertising.