Dr. Sajal Halder | Personalized Recommendation | Best Research Award

Research Fellow at Personalized Recommendation, Charles Sturt University, Australia

👨‍🎓He remarkable academic journey, extensive research contributions, and dedication to the field of psychology are truly commendable. Your wealth of knowledge and diverse skill set reflect a deep commitment to understanding and addressing critical issues such as bullying, inclusion, and socialization.

🔬 He successful completion of a PhD in Psychology, along with the numerous advanced courses and workshops, showcases your continuous pursuit of excellence and expertise in your field.

🏆 The awards and recognitions, including the First Place in the Poster Award at the University of Stavanger, underscore the impact of your research and the high regard it holds in the academic community.

Professional Profiles:

• Google Scholar Profile
• Scopus profile
• ORCID Profile
• LinkedIn Profile

Education:

Dr. Sajal Halder is a dedicated scholar with a strong academic background and a proven track record of research excellence. He earned his Doctor of Philosophy (PhD) in December 2022 from the School of Computing Technologies at the Royal Melbourne Institute of Technology (RMIT) University in Melbourne, Victoria, Australia. His doctoral thesis focused on “Itinerary Recommendation based on Deep Learning” and was supervised by A/Prof Jeffrey Chan and Prof. Xiuzhen Zhang. Prior to his doctoral studies, Dr. Halder completed his Master of Engineering in Computer Engineering at Kyung Hee University, South Korea, graduating in August 2013 with an outstanding CGPA of 4.23/4.30 (equivalent to 95.25%). His master’s thesis, supervised by Prof. Young-Koo Lee, delved into “Supergraph-based Periodic Behaviors Mining in Dynamic Social Networks,” showcasing his expertise in advanced computational techniques. Dr. Halder’s academic journey began with a Bachelor of Science in Computer Science and Engineering from the University of Dhaka, Bangladesh, where he graduated in November 2010 with a commendable CGPA of 3.60/4.00, securing the 5th position out of 59 students. His final project, supervised by Dr. Ashis Kumar Biswas, focused on the “Classification of Multiple Protein Sequences by means of Irredundant Patterns,” demonstrating his early interest in computational biology. Throughout his academic career, Dr. Halder has consistently excelled, as evidenced by his exemplary performance in his Higher Secondary School Certificate (H.S.C) and Secondary School Certificate (S.S.C), where he achieved GPAs of 4.50/5.00 in both, specializing in Science under the Dhaka Board in 2005 and 2003, respectively. His strong foundation in science and technology has laid the groundwork for his successful career in academia and research.

Experience:

Sajal Halder is currently serving as a Research Fellow at Charles Sturt University, located in Wagga Wagga, NSW, Australia, a position held since December 2022. In this role, he has been involved in groundbreaking research, notably in the development of a metadata-based model for detecting malicious and benign packages within the NPM repository. The model introduces two sets of features, namely easy to manipulate (ETM) and difficult to manipulate (DTM), with DTM manipulation relying on long-term planning and monotonic properties. The team has verified the effectiveness of their feature selection using four well-known machine learning techniques and one deep learning technique. Additionally, they have analyzed algorithm performance using metadata manipulation and have recommended improved metadata adversarial attack-resistant algorithms. The experimental analysis conducted on their proposed model has shown a significant reduction of 97.56% in False Positive cases and 80.35% in False Negative cases. Notable achievements of Sajal Halder include his work on “Malicious Package Detection using Metadata Information,” currently in submission for presentation at an A* Conference, and “Install Time Malicious Package Detection on NPM Repository,” which is currently in progress.

Skills:

Sajal Halder possesses a diverse skill set, including expertise in Feature Engineering, Data Scraping, Open Source Software, Machine Learning Model development, Deep Learning techniques, Python programming, Report and Article Writing, Data Visualization, and Teamwork. These skills reflect his proficiency in various aspects of data analysis, from extracting and engineering features to building and evaluating machine learning and deep learning models. His ability to work with open-source software and programming languages like Python demonstrates his adaptability and commitment to leveraging cutting-edge tools for effective data analysis. Furthermore, his proficiency in communicating findings through reports and articles highlights his capability to articulate complex technical concepts effectively. Additionally, his skills in data visualization indicate his capability to present data insights in a visually compelling and informative manner, essential for conveying findings to diverse audiences. Finally, his experience in teamwork underscores his ability to collaborate effectively with others, an important asset in any research or professional environment.

Research/Project Experience:

During his tenure as a Supervisor, Sajal Halder has led several impactful projects, including “Design and Implementation of a Basic Framework for Big Data Analytics,” funded by the ICT Ministry, Government of Bangladesh, from July 1, 2017, to May 31, 2018. This project aimed to establish a foundational framework for conducting Big Data Analytics, addressing the burgeoning need for efficient data processing and analysis in contemporary data-driven environments. Additionally, Mr. Halder contributed to the project “Designing an Efficient Technique for Mining Periodic Patterns in Time Series Databases,” funded by the Ministry of Science and Technology, Government of Bangladesh, running from January 1, 2018, to June 30, 2018. This initiative focused on developing innovative methods for extracting meaningful periodic patterns from time series databases, enhancing the efficiency of data mining processes. Furthermore, he participated in the project “Efficient Spatiotemporal Pattern Mining in Time Series Databases,” supported by the Jagannath University Innovation Fund, spanning from December 1, 2017, to May 31, 2018. This project aimed to advance the field of spatiotemporal pattern mining, addressing the complexities of analyzing time series data with spatial and temporal dimensions. Additionally, in a co-supervisory role, Mr. Halder contributed to the project “Efficient Anomaly Detection Technique on Time Series Graph Data,” funded by the Jagannath University Innovation Fund, operating from December 1, 2017, to May 31, 2018. This project focused on developing novel anomaly detection techniques tailored to time series graph data, enhancing the accuracy and efficiency of anomaly detection processes.

Publications:

Predicting students yearly performance using neural network: A case study of BSMRSTU

• Published in Energy in 2016 with 56 citations.

Movie recommendation system based on movie swarm

• Published in Energy in 2012 with 35 citations.

Supergraph based periodic pattern mining in dynamic social networks

• Published in Energy in 2017 with 34 citations.

An efficient hybrid system for anomaly detection in social networks

• Published in Energy in 2021 with 31 citations.

Exploring significant heart disease factors based on semi supervised learning algorithms

• Published in Energy in 2018 with 25 citations.

Smart disaster notification system

• Published in Energy in 2017 with 23 citations.

Transformer-based multi-task learning for queuing time aware next poi recommendation

• Published in Energy in 2021 with 20 citations.

Smart CDSS: Integration of social media and interaction engine (SMIE) in healthcare for chronic disease patients

• Published in Energy in 2015 with 20 citations.

Link prediction by correlation on social network

• Published in Energy in 2017 with 16 citations.