Gaurav Gupta

Bio: Gaurav Gupta is an academic researcher from Maharaja Surajmal Institute of Technology. The author has contributed to research in topics: Support vector machine & Structured support vector machine. The author has an hindex of 1, co-authored 1 publications receiving 4 citations.

Performance comparison of Support Vector Regression and Relevance Vector Regression for facial expression recognition

Abstract: This paper compares the performance of Relevance Vector Regression and Support Vector Regression for the purpose of facial expression recognition. The Support Vector Machine (SVM) is a state-of-the-art technique for regression and classification, but lacks the probabilistic treatment which is overcome by Relevance Vector Machine (RVM). Though SVM's have a good generalization performance, but their results are in general less sparse. This sometimes results in almost all of the training data to be used as Support Vectors. Comparing with RVM, the results obtained are relatively more sparse than SVM which results in lesser number of Relevance Vectors ultimately leading to lesser computation overhead. The above models are compared for facial expression recognition on Cohn Kanade database. Local Binary Pattern features are extracted from facial images. These are preprocessed for illumination and size, and also for dimensionality reduction before being used for training the RVM and SVM models. The paper concludes with a comparison of the SVM and RVM on the basis of test results.

Meteorology Visibility Estimation by Using Multi-Support Vector Regression Method

Abstract: Meteorological visibility measures the transparency of the atmosphere or air and it provides important information for road, flight and sea transportation safety. Problem of pollution can also affect the visibility of a certain area. Measurement and estimation of visibility is a challenging and complex problem as visibility is affected by various factors such as dust, smoke, fog and haze. Traditional digital image-based approach for visibility estimation involve applications of the meteorology law and mathematical analysis. Digital image-based and machine learning approach can be one of the solutions to this complex problem. In this paper, we propose an intelligent digital method for visibility estimation. Effective regions are first extracted from the digital images and then classified into different classes by using Support Vector Machines (SVM). Multi-Supported Vector Regression (MSVR) models are used to predict the meteorological visibility by using the image features values generated by VGG Neural Network. SVR machine learning method is used for model training and the resulting system can be used for meteorological visibility estimation.

Relative Analysis of ML Algorithm QDA, LR and SVM for Credit Card Fraud Detection Dataset

Abstract: The task is to detect fraudulent online purchases such that consumers of credit card companies are not compensated for purchases they have also not purchased. Several machine learning is techniques were used to identify the fraud detection and accuracy of fraud detections has varied based on the data and model design, In this paper, having been worked on evaluating and pre-processing data sources and perhaps even the development of several feature extraction algorithms such as Quadratic Discriminant Analysis (QDA), Logistic Regression (LR) and Support Vector Machine (SVM).

Microorganism Estimation in a Shrimp Pond Using Gaussian Process Regressor and Gradient Tree Boosting

Abstract: The balance of the aquatic ecosystem is an influential factor in the world of aquaculture, especially in shrimp cultivation. The one that plays a role in that ecosystem is aquatic microorganisms such as vibrio, bacteria, and algae. Therefore, farmers need to know their number and ratio to maintain the shrimp growth. Thus, in this research, models that can estimate vibrio-bacteria ratio and number of algae are developed. These models are formed from aquaculture datasets which are modeled using machine learning algorithms named Gaussian process regressor (GPR) and gradient tree boosting (GTB). Other processing techniques like data pre-processing, feature decomposition, and optimization are also applied to improve model performance. Moreover, these models are also compared to other models which are modeled using another machine learning algorithm like support vector regression (SVR), Lasso, and kernel ridge regression (KRR), so that the best models can be determined. Based on k-fold cross-validation, the GPR model has the best performance in estimating the vibrio-bacteria ratio with mean absolute error (MAE) value of 0.02482 and explained variance score of 0.96515. Then, in the algae estimation, the best performance is achieved by the GTB model with MAE value of 6.55554 and explained variance score of 0.33001.

Machine Learning-Based Imputation Approach with Dynamic Feature Extraction for Wireless RAN Performance Data Preprocessing

Abstract: Machine learning (ML) in wireless mobile communication is becoming more and more customary, with application trends leaning toward performance improvement and network automation. The radio access network (RAN), critical for service access, frequently generates performance data that mobile network operators (MNOs) and researchers leverage for planning, self-optimization, and intelligent network operations. However, missing values in the RAN performance data, as in any valuable data, impact analysis. Poor handling of such missing data in the RAN can distort the relationships between different metrics, leading to inaccurate and unreliable conclusions and predictions. Therefore, there is a need for imputation methods that preserve the overall structure of the RAN data to an optimal level. In this study, we present an imputation approach for handling RAN performance missing data based on machine learning algorithms. The method customizes the feature-extraction mechanism by using dynamic correlation analysis. We apply the method to actual RAN performance indicator data to evaluate its performance. We finally compare and evaluate the proposed approach with statistical imputation techniques such as the mean, median, and mode. The results show that machine learning-based imputation, as approached in this experimental study, preserves some relationships between KPIs compared to non-ML techniques. Random Forest regressor gave the best performance in imputing the data.