Abstract Learning is a lifelong process in this competitive world where we have to update the knowledge everyday with global exposure. To sustain competitiveness in education we induce technology in to learning process. Using the tools of technology, learning and teaching process happens in any electronic medium such as audio, video, TV and satellites across the world. E-learning is technology enabled learning... Learning is a lifelong process in this competitive world where we have to update the knowledge everyday with global exposure. To sustain competitiveness in education we induce technology in to learning process. Using the tools of technology, learning and teaching process happens in any electronic medium such as audio, video, TV and satellites across the world. E-learning is technology enabled learning method which emerged as a potential tool for education industry and also being used for training and development in many other industrial sectors. Higher education is at booming stage, where people around the world spend more than 2 trillion USD for education. E-learning plays a major role in knowledge dissemination where more than 90% of colleges in the United States offer e-learning courses and countries like India have a huge potential growth for higher education through e learning mode. There is an existing gap in student enrolment (CAGR) and faculty Compound Annual Growth Rate (CAGR). It is also evident that a huge gap exists between the undergraduate and post graduate enrolments in India. As per population census more than 60% of people in India are from middle age group who are engaged in full time employment and would want to enrich and update their knowledge with out attending class room teaching. The biggest challenge the E-learning faces is monitoring and validity of the person as to the course has been taken and gone through the assessment in a legal or right manner. The higher education institutions and corporate agree upon making E-learning a part of the course but do not consider it as equivalent to class room teaching. This study is about the readiness of the students to e-learning process and usage. It analyzes the levels of awareness of the student, degree of acceptance and adoption of the student towards the e-learning environment. E-learning courses provide global exposure and help students to have real time experience even in distance education mode. This study tries to identify the demographic factors that influence on three important variables like the readiness of the student towards e-learning environment and knowledge and level of comfort in technology they use. Read More Read Less

Abstract Clustering finds variety of application in a wide range of disciplines because it is mostly helpful for grouping of similar data objects together. Due to the wide applicability, different algorithms have been presented in the literature for segmenting large multidimensional data into discernible representative clusters. Accordingly, in this paper, Kernel-based exponential grey wolf optimizer is developed for rapid centroid estimation in data clustering.... Clustering finds variety of application in a wide range of disciplines because it is mostly helpful for grouping of similar data objects together. Due to the wide applicability, different algorithms have been presented in the literature for segmenting large multidimensional data into discernible representative clusters. Accordingly, in this paper, Kernel-based exponential grey wolf optimizer is developed for rapid centroid estimation in data clustering. Here, a new algorithm, called (EGWO) is newly proposed to search the cluster centroids with a new objective evaluation which considered two parameters called logarithmic kernel function and distance difference between two top clusters. Based on the new objective function and the modified EGWO algorithm, centroids are encoded as position vectors and the optimal location is found for the final clustering. The proposed EGWO algorithm is evaluated with banknote authentication Data Set and iris dataset using four metrics such as, MSE, F-measure, Rand co-efficient and jaccord coefficient. From the outcome, we proved that the proposed EGWO algorithm outperformed the existing PSC, mPSC and GWO algorithm. Read More Read Less

Abstract Recently many machine learning based AI systems have been designed as black boxes. These are the systems that hide the internal logic from the users. Lack of transparency in decision making limits their use in various real world applications. In this paper, we propose a framework that utilizes formal concept analysis to explain AI models. We use classification analysis to study... Recently many machine learning based AI systems have been designed as black boxes. These are the systems that hide the internal logic from the users. Lack of transparency in decision making limits their use in various real world applications. In this paper, we propose a framework that utilizes formal concept analysis to explain AI models. We use classification analysis to study abnormalities in the data which is further used to explain the outcome of machine learning model. The ML method used to demonstrate the ideas is two class classification problem. We validate the proposed framework using a real world machine learning task: diabetes prediction. Our results show that using a formal concept analysis approach can result in better explanations. Read More Read Less

Abstract In this work, we described how to validate the soundness of textual explanations in a domain-independent scenario. Then we proceed to find how the users understand the descriptions and which are the factors determining whether explanations are acceptable. This work found that the discourse tree (complete DT) determines the acceptability of explanation. A complete DT is a sum of a traditional DT for... In this work, we described how to validate the soundness of textual explanations in a domain-independent scenario. Then we proceed to find how the users understand the descriptions and which are the factors determining whether explanations are acceptable. This work found that the discourse tree (complete DT) determines the acceptability of explanation. A complete DT is a sum of a traditional DT for a paragraph of actual text and an imaginary DT for a text about entities used but not explicitly defined in the actual text. Read More Read Less