Srivani Prabhakara
Srivani Prabhakara has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Durga Bhavani Attluri
Visvesvaraya Technological University
2 shared publications A Niranjan
Department of CSE Dayananda Sagar University Kudlu Gate Bangalore Karnataka India
1 shared publicationB. Tamilselvi
Vel Tech High Tech Dr. Rangarajan Dr. Sakunthala Engineering College, India
1 shared publicationC. R. Yamuna Devi
Dr. Ambedkar Institute of Technology
1 shared publicationDevi C R Yamuna
1 shared publication- Elangovan Muniyandy
Department of Biosciences, Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences, Chennai, India
1 shared publication
Latest Publications
- 23 July 2025
- Published by Springer Nature
- p. 279-294
Software Effort Estimation Using AIML and a User Interface
- 23 July 2025
- Published by Springer Nature
- p. 279-294
Software Effort Estimation Using AIML and a User Interface
- 1 May 2025
- Published by IGI Global Scientific Publishing
- p. 359-374
NLP Voice Assistance for IoT Autonomous Vehicles Using ML Algorithms for Seamless Navigation
The goal of this research project is to ascertain whether or not combining Internet of Things (IoT) and Natural Language Processing (NLP) technology could improve the caliber of voice-assisted navigation in self-driving cars. The system under consideration utilizes machine learning techniques to provide smooth navigation and an easy-to-understand voice-based interface for car administration. The preprocessing step includes speech-to-text conversion, which is the process of converting spoken commands into text for additional analysis. The preparatory processing step is another name for this stage. Utilizing time-series data analysis is essential to completing the feature selection process. Finding important patterns that are necessary for navigational judgments requires a study of the vehicle's sensor data, which includes GPS, speed, and ambient inputs. To find important trends, this analysis is required. Sequence models—more especially, Recurrent Neural Networks (RNNs) or Long Short-Term Memory (LSTM) networks—are used during the classification process.
- 1 May 2025
- Published by IGI Global Scientific Publishing
- p. 359-374
NLP Voice Assistance for IoT Autonomous Vehicles Using ML Algorithms for Seamless Navigation
The goal of this research project is to ascertain whether or not combining Internet of Things (IoT) and Natural Language Processing (NLP) technology could improve the caliber of voice-assisted navigation in self-driving cars. The system under consideration utilizes machine learning techniques to provide smooth navigation and an easy-to-understand voice-based interface for car administration. The preprocessing step includes speech-to-text conversion, which is the process of converting spoken commands into text for additional analysis. The preparatory processing step is another name for this stage. Utilizing time-series data analysis is essential to completing the feature selection process. Finding important patterns that are necessary for navigational judgments requires a study of the vehicle's sensor data, which includes GPS, speed, and ambient inputs. To find important trends, this analysis is required. Sequence models—more especially, Recurrent Neural Networks (RNNs) or Long Short-Term Memory (LSTM) networks—are used during the classification process.
- Open Access
- 1 October 2024
- Published by Institute of Advanced Engineering and Science in International Journal of Electrical and Computer Engineering (IJECE)
- Vol. 14 (5) , 5543-5553
Conflict-driven learning scheme for multi-agent based intrusion detection in internet of things
This paper introduces an effective intrusion detection system (IDS) for the internet of things (IoT) that employs a conflict-driven learning model within a multi-agent architecture to enhance network security. A double deep Q-network (DDQN) reinforcement learning algorithm is implemented in the proposed IDS with two specialized agents, the defender and the challenger. These agents engaged in an antagonistic adaptation process that dynamically refined their strategies through continual interaction within a custom-made environment designed using OpenAI Gym. The defender agent aims to identify and mitigate threats by matching the actions of the challenger agent, which is designed to simulate potential attacks in the environment. The study introduces a binary reward mechanism to encourage both agents to explore and exploit different actions and discover new strategies as a response to adversarial actions. The results showcase the effectiveness of the proposed IDS in terms of higher detection rate the comparative analysis also validates the effectiveness of the proposed IDS scheme with an accuracy of approximately 96%, outperforming similar existing approaches.
- 1 October 2024
- Published by Institute of Advanced Engineering and Science in International Journal of Electrical and Computer Engineering (IJECE)
- Vol. 14 (5) , 5543-5553
Open AccessConflict-driven learning scheme for multi-agent based intrusion detection in internet of things
This paper introduces an effective intrusion detection system (IDS) for the internet of things (IoT) that employs a conflict-driven learning model within a multi-agent architecture to enhance network security. A double deep Q-network (DDQN) reinforcement learning algorithm is implemented in the proposed IDS with two specialized agents, the defender and the challenger. These agents engaged in an antagonistic adaptation process that dynamically refined their strategies through continual interaction within a custom-made environment designed using OpenAI Gym. The defender agent aims to identify and mitigate threats by matching the actions of the challenger agent, which is designed to simulate potential attacks in the environment. The study introduces a binary reward mechanism to encourage both agents to explore and exploit different actions and discover new strategies as a response to adversarial actions. The results showcase the effectiveness of the proposed IDS in terms of higher detection rate the comparative analysis also validates the effectiveness of the proposed IDS scheme with an accuracy of approximately 96%, outperforming similar existing approaches.
- 23 September 2023
- Published by Springer Nature
- p. 559-566
An Internet of Things-Based Smart Asthma Inhaler Integrated with Mobile Application
- 23 September 2023
- Published by Springer Nature
- p. 559-566
An Internet of Things-Based Smart Asthma Inhaler Integrated with Mobile Application
- Open Access
Contextual Cluster-Based Glow-Worm Swarm Optimization (GSO) Coupled Wireless Sensor Networks for Smart Cities
The cluster technique involves the creation of clusters and the selection of a cluster head (CH), which connects sensor nodes, known as cluster members (CM), to the CH. The CH receives data from the CM and collects data from sensor nodes, removing unnecessary data to conserve energy. It compresses the data and transmits them to base stations through multi-hop to reduce network load. Since CMs only communicate with their CH and have a limited range, they avoid redundant information. However, the CH’s routing, compression, and aggregation functions consume power quickly compared to other protocols, like TPGF, LQEAR, MPRM, and P-LQCLR. To address energy usage in wireless sensor networks (WSNs), heterogeneous high-power nodes (HPN) are used to balance energy consumption. CHs close to the base station require effective algorithms for improvement. The cluster-based glow-worm optimization technique utilizes random clustering, distributed cluster leader selection, and link-based routing. The cluster head routes data to the next group leader, balancing energy utilization in the WSN. This algorithm reduces energy consumption through multi-hop communication, cluster construction, and cluster head election. The glow-worm optimization technique allows for faster convergence and improved multi-parameter selection. By combining these methods, a new routing scheme is proposed to extend the network’s lifetime and balance energy in various environments. However, the proposed model consumes more energy than TPGF, and other protocols for packets with 0 or 1 retransmission count in a 260-node network. This is mainly due to the short INFO packets during the neighbor discovery period and the increased hop count of the proposed derived pathways. Herein, simulations are conducted to evaluate the technique’s throughput and energy efficiency.
Open AccessContextual Cluster-Based Glow-Worm Swarm Optimization (GSO) Coupled Wireless Sensor Networks for Smart Cities
The cluster technique involves the creation of clusters and the selection of a cluster head (CH), which connects sensor nodes, known as cluster members (CM), to the CH. The CH receives data from the CM and collects data from sensor nodes, removing unnecessary data to conserve energy. It compresses the data and transmits them to base stations through multi-hop to reduce network load. Since CMs only communicate with their CH and have a limited range, they avoid redundant information. However, the CH’s routing, compression, and aggregation functions consume power quickly compared to other protocols, like TPGF, LQEAR, MPRM, and P-LQCLR. To address energy usage in wireless sensor networks (WSNs), heterogeneous high-power nodes (HPN) are used to balance energy consumption. CHs close to the base station require effective algorithms for improvement. The cluster-based glow-worm optimization technique utilizes random clustering, distributed cluster leader selection, and link-based routing. The cluster head routes data to the next group leader, balancing energy utilization in the WSN. This algorithm reduces energy consumption through multi-hop communication, cluster construction, and cluster head election. The glow-worm optimization technique allows for faster convergence and improved multi-parameter selection. By combining these methods, a new routing scheme is proposed to extend the network’s lifetime and balance energy in various environments. However, the proposed model consumes more energy than TPGF, and other protocols for packets with 0 or 1 retransmission count in a 260-node network. This is mainly due to the short INFO packets during the neighbor discovery period and the increased hop count of the proposed derived pathways. Herein, simulations are conducted to evaluate the technique’s throughput and energy efficiency.
- Open Access
- 12 December 2022
- Published by Indian Society for Education and Environment in Indian Journal of Science and Technology
- Vol. 15 (46) , 2527-2533
Monitoring Environment Parameters of Gerbera Flower Cultivation in Greenhouse using Internet of Things
Objectives: To propose a system based on monitoring and analyzing of greenhouse environment, built for cultivating Gerbera with advanced technologies like the Internet of Things(IoT), Android applications, and cloud. Methods: To ensure optimal growth of the plant in a greenhouse, smart applications with sensors and Mobile apps are implemented for remote monitoring. Sensors are attached to the Arduino interface which collects and broadcasts the data onto the IoT cloud. This cloud platform (Ubidots) creates real-time dashboards for user-friendly graphic visualization. Further, through the Android Mobile application interface, the data can be regularly monitored and tracked by the end user to take necessary actions. Findings: The system is designed to automatically monitor four environmental parameters such as temperature, humidity, soil moisture, and light inside the greenhouse through IoT. The data recorded in the cloud is visualized through an app and the user can monitor remotely the evident conditions of greenhouse parameters. If any of these parameter values crosses the threshold set, then the notification will be sent to the end user. A mobile application developed will guide the farmers to remotely monitor the greenhouse and influence the grower’s decisions. Novelty: The proposed work ensures that the quality and yield of crops can be effective with automatic remote monitoring with the help of technology. This work highlights designing a user-friendly app for farmers which is safe and can track the current conditions, detect abnormalities, and easy water management. The Signup/login page, data visualization, and alert messages are implemented in the app which marks farming as more efficient compared to traditional methods. Keywords: Gerbera; Greenhouse; Ubidots; Internet of Things; Android
- 12 December 2022
- Published by Indian Society for Education and Environment in Indian Journal of Science and Technology
- Vol. 15 (46) , 2527-2533
Open AccessMonitoring Environment Parameters of Gerbera Flower Cultivation in Greenhouse using Internet of Things
Objectives: To propose a system based on monitoring and analyzing of greenhouse environment, built for cultivating Gerbera with advanced technologies like the Internet of Things(IoT), Android applications, and cloud. Methods: To ensure optimal growth of the plant in a greenhouse, smart applications with sensors and Mobile apps are implemented for remote monitoring. Sensors are attached to the Arduino interface which collects and broadcasts the data onto the IoT cloud. This cloud platform (Ubidots) creates real-time dashboards for user-friendly graphic visualization. Further, through the Android Mobile application interface, the data can be regularly monitored and tracked by the end user to take necessary actions. Findings: The system is designed to automatically monitor four environmental parameters such as temperature, humidity, soil moisture, and light inside the greenhouse through IoT. The data recorded in the cloud is visualized through an app and the user can monitor remotely the evident conditions of greenhouse parameters. If any of these parameter values crosses the threshold set, then the notification will be sent to the end user. A mobile application developed will guide the farmers to remotely monitor the greenhouse and influence the grower’s decisions. Novelty: The proposed work ensures that the quality and yield of crops can be effective with automatic remote monitoring with the help of technology. This work highlights designing a user-friendly app for farmers which is safe and can track the current conditions, detect abnormalities, and easy water management. The Signup/login page, data visualization, and alert messages are implemented in the app which marks farming as more efficient compared to traditional methods. Keywords: Gerbera; Greenhouse; Ubidots; Internet of Things; Android