KCG College of Technology

About: KCG College of Technology is a based out in . It is known for research contribution in the topics: Computer science & The Internet. The organization has 427 authors who have published 381 publications receiving 2193 citations.

Adaptive cognitive intelligence in analyzing employee feedback using LSTM

Abstract: Cognitive computing is the mirroring of human brain and this is made possible by using natural language processing, pattern recognition and data mining. By mirroring the human brain (Cognitive computing system), helps to solve some of the complicated problems without much of human supervision. In the fast-changing world, the major challenge every organization facing is difficulty in retaining its employees. Employees may leave an organization due to low salary, overwork, lack of opportunities and recognition, work culture, work-life imbalance etc. Better ways to retain employees is to understand their requirements and fulfill them. The proposed employee feedback sentiment analysis system collects the employee feedback reviews from open forums and perform sentiment analysis using Recurrent Neural Network – Long Short-term Memory (RNN-LSTM) algorithm. On performing Sentiment analysis, employee review comments are classified as Positive or Negative. A report is generated and sent to the HR of the organization as webapp or mobile app. The report has total number of positive and negative comments and positive and negative counts with respect to salary, work pressure etc. With the report, the organization can arrive at identifying social sentiments of their brand and may take corrective actions to retain employees which benefits both organization and employees. This paper also captures the performance of various models in training and predicting the employee feedback dataset and the models evaluated are Logistic Regression, Support Vector Machine, Random Forest Classifier, AdaBoost Classifier, Gradient Boosting Classifier, Decision Tree Classifier and Gaussian Naïve Bayes. The classification report and accuracy of each model is captured. The dataset size was gradually increased from 200 to 1000 and accuracy was predicted for each model. It was identified that the accuracy of machine learning algorithms was ranging between 66% to 85%. On training RNN-LSTM algorithm with dataset of size 30 k, the accuracy was 88%. It was identified that Deep learning algorithm RNN-LSTM performs better with huge dataset. Increasing dataset size still increase the performance of RNN-LSTM algorithm in training and prediction. Thus, the objective function of the proposed model to perform sentiment analysis on employee feedback review comments is achieved successfully.

Design and implementation of hybrid integration of cognitive learning and chaotic countermeasures for side channel attacks

Abstract: Security in embedded systems is considered to be more important and needs to be a diagnosis for every minute. Also with the advent of the Internet of Things (IoT), security in the embedded system has reached its new peak of dimension. A Mathematically secure algorithm was formulated and runs on the cryptographic chips which are embedded in the systems, but secret keys can be at risk and even information can be retrieved by the prominent side-channel attacks. Fixed encryption keys, non-intelligent detection of side-channel attacks are some of the real-time challenges in an existing system of encryption. Following the limitations of existing systems, this research article focuses on the integration of powerful machine learning algorithms by retrieving the secret key information with countermeasures methodology using the chaotic logistic maps and includes the following contributions: (a) Preparation of Data Sets from the Power consumption traces captured from ARTIX-7 FPGA boards while running the Elliptical Curve Cryptography(ECC) on it (b) Implementation of High Speed and High Accurate Single feed-forward learning machines for the detection and classification of side-channel attacks (c) Design of Chaotic Countermeasures using 3-Dlogistic maps for attacked bits. The test_bed has been developed using the integration of FPGA along with Cortex-A57 architectures for experimentation of the proposed work and various evaluation parameters such as Accuracy, F-calls, Precision rates, sensitivity, and correlation co-efficient, entropy were calculated and analyzed. Moreover, the parameters of the proposed system which has been analyzed prove to outperform the other existing algorithms in terms of performance and detection.

Detection of Pulmonary Nodules Using Thresholding and Fractal Analysis

Abstract: Automated detection of pulmonary nodules helps radiologists in early detection of lung cancer from computed tomography (CT) scans. It is very costly computationally because of its complexity of the process. The CT scan has more advantages than other computational algorithms. The preprocessed CT scan is thresholded using Otsu’s method and the lung region is segmented using K-means Clustering which is based on geometric features. Texture based feature analysis algorithm is used to identify the major descriptors. The Artificial Neural Networks (ANN) is used for training, testing and validation process takes place to identify nodule and classify in stages i.e. Stage 1 (initial), Stage 2 (middle) and Stage 3 (critical). The results obtained in this method has been checked for accuracy.

Low rate DDoS mitigation using real-time multi threshold traffic monitoring system

Abstract: The low rate distributed denial of service (DDoS) attack has been identified as most vulnerable to the network services which has been studied recently. The approaches consider only the high rate DoS attacks and ignore rest in low rate. The existing techniques suffer with poor detection of low rate attacks as they consider only limited features of network traffic. Variety of techniques mitigate such threats using different parameters like amount of data in service packet as payload, number of intermediate nodes, and so on. The previous techniques struggle to detect and mitigate them in efficient way. Towards improving the detection and mitigation performance of low rate threats, the author presents a novel real time traffic monitoring algorithm which uses multi threshold traffic analysis. By considering the payload, hop count, latency, packet counts, the method analyzes the real time traffic. Using the features obtained from the traffic, the method computes the low rate threat measure. Based on computed threat measure, the packets trustworthy have been validated. The method produces higher detection rate in low rate DDoS attack detection and produces efficient results.

Non-linear intermittent instabilities and their control in an interleaved DC/DC converter

Abstract: The undesirable intermittent instabilities in periodically driven parallel power converter are caused due to periodic interferences in input voltage This is a more common problem in power electronic circuit design and such operations are usually avoided by controlling the circuit parameters In this study, the mechanism and conditions for the emergence of intermittent instabilities and remerging chaotic band attractors (or Feigenbaum sequences) in a master-slave controlled parallel buck converter are investigated It is found that sinusoidal-type interference in input voltage at frequencies near the switching frequency or its rational multiples of this circuit results typically in period-bubbling, chaos and intermittency It is also shown that an optimal, phase-shifted sinusoidal interference added to the reference voltage controls the period-bubbling behaviour and significantly extends the parameter range of desirable period-1 operation The dynamics of this converter has been first investigated with suitable numerical simulations The ordered and the chaotic dynamics of this system have been further mathematically described and analysed with a simple discrete map and experimental means Experimental observations are found to be in good agreement with the analytical and simulation results