CALL US
(0343)-2501353,2504106
Mob: +91-6297128554 | Email: info@bcrec.ac.in
CALL US
|
01-07-2010 |
|
|
IT |
|
|
186727 |
|
|
AB+ |
|
|
HINDUISM |
|
|
1-449846821 |
|
|
01-07-2006 |
| Teaching | Research | Industry |
|---|---|---|
| 228 | 13 | 0 |
| Date | Title | Conference | DOI | Link |
|---|---|---|---|---|
| 23-03-2026 | Evaluating Human AI Collaboration Through Survey Based Random Forest Approach | 2025 International Conference on Signal Processing, Computation, Electronics, Power and Telecommunication (IConSCEPT) | DOI | View |
| 18-06-2025 | Predictive Framework for Sustainable Engineering through Machine Learning and Cross-Sector Collaboration | First International Conference on Engineering and Technology for a Sustainable Future (ICETSF-2025) | DOI | View |
| 18-04-2025 | Breast Cancer Classification Using Machine Learning: A Comprehensive Review and Analysis | Smart Systems and Wireless Communication | DOI | View |
| 28-06-2024 | Exploring the Relationship Between Rent and Flat Prices Through Random Forest and Grid Search | 2024 International Conference on Circuit, Systems and Communication (ICCSC) | DOI | View |
| 23-03-2023 | Disease Detection in Paddy Crop using Machine Learning Techniques | 2023 International Conference on Sustainable Computing and Data Communication Systems (ICSCDS) | DOI | View |
| 10-02-2023 | Comparing the Use of Short Video Sharing Applications for Optimizing User Engagement | 2023 IEEE 3rd International Conference on Technology, Engineering, Management for Societal impact using Marketing, Entrepreneurship and Talent (TEMSMET) | DOI | View |
| 21-02-2015 | Proposed Artificial Intelligence Based Authentication of User in Remote System | International Science Congress Association (ISCA), Recent trends in Computations and Mathematical Analysis in Engineering and Sciences( CRCMAS 2015 ) | DOI | View |
| Committee Name | Start Date | End Date |
|---|---|---|
| ACADEMIC AUDIT COMMITTEE_IT | 07-09-2022 | 03-06-2026 |
| DEPARTMENTAL MAGAZINE AND ACTIVITY COMMITTEE | 07-09-2022 | 03-06-2026 |
| DEPARTMENTAL ACADEMIC COUNCIL | 01-07-2017 | 03-06-2026 |
| Title | Patent Number | Patent Office | Date | Status | Country | Application No. | Abstract | Expiration Date |
|---|---|---|---|---|---|---|---|---|
| SENTIMENT-AWARE STOCK TRADING BOT USING REINFORCEMENT LEARNING | Chennai | 05-09-2025 | PUBLISHED | INDIA | 202541075069 | The invention presents a sentiment-aware automated trading framework that fuses real-time sentiment interpretation with a reinforcement-learning trading engine. Central to the architecture is a sentiment extraction module that generates sentiment scores from diverse unstructured sources—namely financial news articles, social media feeds, and thematic blogs. Concomitantly, a market data unit collects conventional technical indicators alongside price time series. The resultant sentiment and market parameters are reconciled within a feature aggregation unit, yielding a coherent market state representation. A reinforcement-learning trading agent then operates upon this representation, selecting among the discrete actions of buying, selling, or holding, with the objective of maximising expected cumulative returns. This objective is qualified by penalties for transaction costs and by the modulation of actions according to current market volatility. A risk management sub-layer monitors exposure and dynamically adjusts position sizing. The architecture accommodates both high-fidelity simulation environments and live trading terminals. By integratively modelling qualitative sentiment alongside quantitative indicators, the presented framework delivers a context-sensitive and robust strategy for algorithmic trading. | 04-09-2030 | |
| Deep Reinforcement Learning for Energy-Efficient Computation Offloading with DVFS for Time-Critical IoT Applications in Edge Computing | IN-UP75020767738770V | DELHI / MUMBAI / CHENNAI / KOLKATA | 01-09-2023 | PUBLISHED | INDIA | 202341046984 | Internet of Things (IoT) is a steadily growing industry. It investigates the infrastructure and protocols that allow large and tiny computers to connect to the Internet, share data, and utilise it. As a consequence of this knowledge, people's interactions with their surroundings are evolving. It lays the groundwork for new applications and services that will streamline and accelerate manual tasks. Everything in this composition is intricately intertwined. The majority of the time, Internet of Things-connected devices generate a great deal of data. These devices previously transmitted data to centralised cloud servers. These servers provided the computing power for Cloud Computing (CC). This strategy has a number of drawbacks, including longer wait times, an increased demand for network speed, concerns about privacy and security, and so on. Edge Computing (EC) complements the potent cloud servers in the data centre. It offers tremendous processing power near to the data source, accelerates data transfer, and maintains privacy. Small, battery-powered Internet of Things devices are a significant problem because they consume a great deal of energy. In recent years, the Internet of Things (IoT) community has become increasingly concerned with its energy consumption. This has resulted in a variety of strategies for reducing energy consumption while meeting the increasing demand for computing capacity. When it comes to energy consumption, the central processing unit (CPU) of the computer ranks first. DVFS, which stands for "dynamic voltage and frequency scaling," is a power-saving method employed by modern computers. In this thesis, we demonstrate how to construct an offloading method for a distributed vector file system (DVFS) using reinforcement learning at the edge. Our objective is to discover ways to increase the electrical efficiency of IoT devices. Experiment results indicate that this method can reduce energy consumption while still conducting the application and service-required computing tasks. The proposed method performs better than the native Linux governors in terms of energy efficiency. The average quantity of energy used decreases by 5% when compared to on-demand governors and by 9.5% when compared to conservative governors. | 12-07-2043 |