Papers by Angel Deborah S
Contextual emotion detection on text using gaussian process and tree based classifiers
Intelligent Data Analysis
It is challenging for machine as well as humans to detect the presence of emotions such as sadnes... more It is challenging for machine as well as humans to detect the presence of emotions such as sadness or disgust in a sentence without adequate knowledge about the context. Contextual emotion detection is a challenging problem in natural language processing. As the use of digital agents have increased in text messaging applications, it is essential for these agents to provide sensible responses to its users. The present work demonstrates the effectiveness of Gaussian process detecting contextual emotions present in a sentence. The results obtained are compared with Decision Tree and ensemble models such as Random Forest, AdaBoost and Gradient Boost. Out of the five models built on a small dataset with class imbalance, it has been found that Gaussian Process classifier predicts emotions better than the other classifiers. Gaussian Process classifier performs better by taking predictive variance into account.
With the increase in consumption of energy and population, there is a grave need to conserve ener... more With the increase in consumption of energy and population, there is a grave need to conserve energy in every way possible. The inability to access and control the appliances from remote locations is one of the major reasons for energy loss. A web or an android application is used by the users to give instructions to these systems. This system can make use of a host of communication methods such as Wi-Fi, GSM, Bluetooth, ZigBee. Different controlling devices and configurations can be found in existing systems. Such systems have been found already in many places for a wide variety of applications. This paper presents a survey of all such systems.
ART-CPN Based Aircraft Navigation by GPS/INS Data Integration
Artificial Intelligent Systems and Machine Learning, 2011
GPS and INS are commonly integrated using a Kalman filter (KF) to provide a robust aircraft navig... more GPS and INS are commonly integrated using a Kalman filter (KF) to provide a robust aircraft navigation solution, overcoming drawbacks of GPS satellite signals blockage. This work presents an alternative method of integrating GPS and INS data, called Artificial Neural Networks. This method uses Adaptive Resonance Theory-Counter Propagation Neural Networks (ART-CPN) to predict the INS position error. The performance of ART-CPN is analyzed using real time data in terms of Root Mean Square Error (RMSE), Performance Index (PI), number of hidden neuron, number of epochs and learning rate. The performance of Forward only Counter Propagation Network (FCPN) and Full Counter Propagation Network (Full CPN) are also analyzed and compared with ART-CPN. ART-CPN is found to have better clustering ability when compared to FCPN and Full CPN. ART-CPN also has better learning ability and network constructing ability when compared to FCPN and Full CPN. It has better learning speed due to its one step l...
Virtual Fashion Mirror
2020 4th International Conference on Computer, Communication and Signal Processing (ICCCSP), 2020
Everyone loves to go shopping for clothes, except for the hassle that is the trial room experienc... more Everyone loves to go shopping for clothes, except for the hassle that is the trial room experience. The Augmented Reality Fashion Display attempts to streamline the shopping process by removing all the troubles of trying on multiple clothes at a small cramped trail room. The product is capable of identifying a person, displaying clothes and accessories like watches on them in real time. This reduces the time spent waiting in queues and the cumbersomeness of trying on clothes that may have been worn by someone before you in the very same trial room. The applications for this technology can expand well beyond the scope of fashion stores and into the realm of fashion design itself enabling fashion designers to create models of their prototype designs and early concept art and test it out virtually before spending materials and resources to physically manifest these designs.
International Journal of Computer Applications, 2016
Teenage alcohol addiction poses a major problem to the wellbeing of the individual as well as the... more Teenage alcohol addiction poses a major problem to the wellbeing of the individual as well as the society. Prevention of this requires identifying the factors causing this addiction. The existing systems mainly rely on decision trees and are able to isolate the factors causing the addiction. The proposed system will be able to predict whether a student with a set of conditions will get addicted to alcohol or not with high accuracy and thereby verify the extent to which the isolated factors are correct. General Terms Data mining, Student characteristics dataset
Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021)
This paper describes the system used for detecting humor in text. The system developed by the tea... more This paper describes the system used for detecting humor in text. The system developed by the team TECHSSN uses binary classification techniques to classify the text. The data undergoes preprocessing and is given to ColBERT (Contextualized Late Interaction over BERT), a modification of Bidirectional Encoder Representations from Transformers (BERT). The model is retrained and the weights are learned for the dataset. This system was developed for the task 7 of the competition, SemEval 2021.
Preventing general traffic v iolation is not easy in roads or urban streets due to the inattentiv... more Preventing general traffic v iolation is not easy in roads or urban streets due to the inattentive drivers. This small violation is one of the leading causes for accidents. In this paper, we present an intelligent traffic violation detection and traffic flow analysis system (TVDTFA) to monitor and measure red light jumping. Th is system is based upon Radio Frequency Identification (RFID) technology for identification of vehicles on the road. Traffic vio lation detection algorithm is based upon the information retrieved from vehicle an d type of signal from t raffic light. The algorithm also extends to calculating and updating the penalty for vehicles jumping red light. The penalty is retrieved from the informat ion publicly availab le in the traffic police forum. Traffic flow analysis is d one by collecting real t ime vehicle count data and using data analysis techniques.
Contextual Emotion Detection in Text Using Ensemble Learning
Emerging Trends in Computing and Expert Technology, 2019
As human beings, it is hard to interpret the presence of emotions such as sadness or disgust in a... more As human beings, it is hard to interpret the presence of emotions such as sadness or disgust in a sentence without the context, and the same ambiguity exists for machines also. Emotion detection from facial expressions and voice modulation easier than emotion detection from text. Contextual emotion detection from text is a challenging problem in text mining. Contextual emotion detection is gaining importance, as people these days are communicating mainly through text messages, to provide emotionally aware responses to the users. This work demonstrates ensemble learning to detect emotions present in a sentence. Ensemble models like Random Forest, Adaboost and Gradient Boosting have been used to detect emotions. Out of the three models, it has been found that Gradient Boosting Classifiers predicts the emotions better than the other two classifiers.
Home Security System for the Hearing Impaired
2021 5th International Conference on Computer, Communication and Signal Processing (ICCCSP)
This paper aims at assisting the hearing impaired by providing innovative solutions to the numero... more This paper aims at assisting the hearing impaired by providing innovative solutions to the numerous challenges experienced by them. These can be someone at the door, someone forcibly entering into the house or an alarm going off. Since about 5 percent of the world population experience hearing loss, it is critical to develop a home security system dedicated to them. The proposed device provides solutions to all problems the hearing-impaired encounter. Home security is made accessible to the Deaf and hearing impaired by including features that are essential and distinct to them. Hence, it can immensely improve the quality of life of the hearing impaired. The system provides surveillance using the Raspberry Pi which can be monitored from an Android application. A knock at the door, a doorbell press, smoke, gas, fire, Carbon Monoxide (CO) and water levels are monitored using the vibration, button, MQ2, MQ6, Flame, MQ7 and water level sensors, following which an alert is sent in case of unfavorable situations. The system also checks constantly for intruders using an ultrasonic sensor and an alert is sent in case of the same. In addition to this, a panic button that sends an SMS is provided to assist the user in case of an emergency. The proposed system can overcome the shortcomings of previous systems by providing unique and efficient solutions to the prevalent and severe issues faced by the hearing impaired.
Proceedings of The 12th International Workshop on Semantic Evaluation, 2018
Sentiment analysis plays an important role in E-commerce. Identifying ironic and sarcastic conten... more Sentiment analysis plays an important role in E-commerce. Identifying ironic and sarcastic content in text plays a vital role in inferring the actual intention of the user, and is necessary to increase the accuracy of sentiment analysis. This paper describes the work on identifying the irony level in twitter texts. The system developed by the SSN MLRG1 team in SemEval-2018 for task 3 (irony detection) uses rule based approach for feature selection and MultiLayer Perceptron (MLP) technique to build the model for multiclass irony classification subtask, which classifies the given text into one of the four class labels.
Proceedings of The 12th International Workshop on Semantic Evaluation, 2018
The system developed by the SSN MLRG1 team for Semeval-2018 task 1 on affect in tweets uses rule ... more The system developed by the SSN MLRG1 team for Semeval-2018 task 1 on affect in tweets uses rule based feature selection and one-hot encoding to generate the input feature vector. Multilayer Perceptron was used to build the model for emotion intensity ordinal classification, sentiment analysis ordinal classification and emotion classfication subtasks. Support Vector Regression was used to build the model for emotion intensity regression and sentiment intensity regression subtasks.
Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017), 2017
The system developed by the SSN MLRG1 team for Semeval-2017 task 5 on fine-grained sentiment anal... more The system developed by the SSN MLRG1 team for Semeval-2017 task 5 on fine-grained sentiment analysis uses Multiple Kernel Gaussian Process for identifying the optimistic and pessimistic sentiments associated with companies and stocks. Since the comments on the same companies and stocks may display different emotions depending on time, their properities like smoothness and periodicity may vary. Our experiments show that while single Kernel Gaussian Process can learn some properties well, Multiple Kernel Gaussian Process are effective in learning the presence of different properties.
IoT Based Smart Waste Management Reporting And Monitoring System
2021 5th International Conference on Computer, Communication and Signal Processing (ICCCSP)
A survey of voice translation methodologies — Acoustic dialect decoder
2016 International Conference on Information Communication and Embedded Systems (ICICES), 2016
Performance comparison of HONNs and FFNNs in GPS and INS integration for vehicular navigation
2011 International Conference on Recent Trends in Information Technology (ICRTIT), 2011
... [2] M.Malleswaran, Dr.V.Vaidehi, S.Angel Deborah, “Data Fusion Using Multi-Layer Feed Forward... more ... [2] M.Malleswaran, Dr.V.Vaidehi, S.Angel Deborah, “Data Fusion Using Multi-Layer Feed Forward Neural Networks for Land Vehicle Navigation”, International Journal of Engineering Science and Technology, vol. ... [11] SN Sivanandam, S.Sumathi, SNDeepa, “Introduction to ...
CNN based GPS/INS data integration using new dynamic learning algorithm
2011 International Conference on Recent Trends in Information Technology (ICRTIT), 2011
... The second hidden neuron is then recruited using the same process. ... Whereas, CNN uses less... more ... The second hidden neuron is then recruited using the same process. ... Whereas, CNN uses lesser number of hidden neurons when compared to BPN, so CNN is said to have optimal topology. ... 2010. [2] Rashad Sharaf and Aboelmagd Noureldin, Sensor Integration for Satellite ...
Integration of INS and GPS using radial basis function neural networks for vehicular navigation
2010 11th International Conference on Control Automation Robotics & Vision, 2010
ABSTRACT Navigation systems used in recent days rely mainly on Kalman filter to fuse data from gl... more ABSTRACT Navigation systems used in recent days rely mainly on Kalman filter to fuse data from global positioning system (GPS) and the inertial navigation system (INS). In common, INS/GPS data fusion provides reliable navigation solution by overcoming drawbacks such as signal blockage for GPS and increase in position errors with time for INS. Kalman filtering INS/GPS integration techniques used in present days have some inadequacies related to the stochastic error models of inertial sensors, immunity to noise, and observability. This paper aims to introduce a new system integration approach for fusing data from INS and GPS utilizing artificial neural networks (ANN). A multi-layer perceptron ANN has been recently suggested to fuse data from INS and differential GPS (DGPS). Though the integrated system using multi-layer perceptron scheme improves the positioning accuracy, it has shortcomings like complexity with respect to the architecture of multi-layer perceptron networks and limitation of online training algorithm to provide real-time capabilities. This paper, therefore, proposes the use of an alternative ANN architecture. The proposed architecture is based on radial basis function (RBF) neural networks, which generally have simpler architecture and faster training procedures than multi-layer perceptron networks. The RBF-ANN module is trained to predict the INS position error and provide accurate positioning of the moving vehicle.
Data Fusion Using Different Activation Functions in Artificial Neural Networks for Vehicular Navigation
Global positioning System (GPS) and Inertial Navigation System (INS) data can be integrated toget... more Global positioning System (GPS) and Inertial Navigation System (INS) data can be integrated together to provide a reliable navigation. GPS/INS data integration provides reliable navigation solutions by overcoming each of their shortcomings, including signal blockage for GPS and increase in position errors with time for INS. This paper aims to provide GPS/INS data integration utilizing Artificial Neural Network (ANN) architecture. This architecture is based on Feed Forward Neural Networks, which generally includes Radial Basis Function (RBF) neural network and Back Propagation neural network (BPN). These are systematic methods for training multi-layer artificial networks. The BPN-ANN and RBF-ANN modules are trained to predict the INS position error and provide accurate positioning of the moving vehicle. This paper also compares performance of theGPS/INS data integration system by using different activation function like Bipolar Sigmoidal Function (BPSF), Binary Sigmoidal Function (BI...
Analysis of Different Associative Memory Neural Network for GPS/INS Data Fusion
International Journal of Computer Applications, 2015
From the last three decades creating human robotic hand replica with enhanced capabilities is of ... more From the last three decades creating human robotic hand replica with enhanced capabilities is of concern and lot of efforts have been put into it. This paper focuses on understanding the different techniques that are used for human robot interaction in robotic hand arm systems. Diversification is stated in areas of human and robotic hand interaction, the degrees of freedom, the grasping ability, number of fingers and materials used for the hand. The flexibility of grasp is compared in terms of Degrees Of Freedom (DOF) and the number of finger end effectors. The controlling method is either through sensor based or gesture controlled or simulation based or pre-defined positions.
Uploads
Papers by Angel Deborah S