Analyzing Linguistic Features for Answer Re-Ranking of Why-Questions

Why-type non-factoid questions are ambiguous and involve variations in their answers. A challenge in returning one appropriate answer to user requires the process of appropriate answer extraction, re-ranking and validation. There are cases where the need is to understand the meaning and context of a document rather than finding exact words involved in question. The paper addresses this problem by exploring lexico-syntactic, semantic and contextual query-dependent features, some of which are based on deep learning frameworks to depict the probability of answer candidate being relevant for the question. The features are weighted by the score returned by ensemble ExtraTreesClassifier according to features importance. An answer re-ranker model is implemented that finds the highest ranked answer comprising largest value of feature similarity between question and answer candidate and thus achieving 0.64 Mean Reciprocal Rank (MRR). Further, answer is validated by matching the answer type of answer candidate and returns the highest ranked answer candidate with matched answer type to a user.

Analyzing Linguistic Features for Answer Re-Ranking of Why-Questions

Why-type non-factoid questions are ambiguous and involve variations in their answers. A challenge in returning one appropriate answer to user requires the process of appropriate answer extraction, re-ranking and validation. There are cases where the need is to understand the meaning and context of a document rather than finding exact words involved in question. The paper addresses this problem by exploring lexico-syntactic, semantic and contextual query-dependent features, some of which are based on deep learning frameworks to depict the probability of answer candidate being relevant for the question. The features are weighted by the score returned by ensemble ExtraTreesClassifier according to features importance. An answer re-ranker model is implemented that finds the highest ranked answer comprising largest value of feature similarity between question and answer candidate and thus achieving 0.64 Mean Reciprocal Rank (MRR). Further, answer is validated by matching the answer type of answer candidate and returns the highest ranked answer candidate with matched answer type to a user.

Review on Different Software Tools for Deep Learning

Deep Learning Applications are being applied in various domains in recent years. Training a deep learning model is a very time consuming task. But, many open source frameworks are available to simplify this task. In this review paper we have discussed the features of some popular open source software tools available for deep learning along with their advantages and disadvantages. Software tools discussed in this paper are Tensorflow, Keras, Pytorch, Microsoft Cognitive Toolkit (CNTK). Keywords: Deep Learning, Frameworks, Open Source, Tensorflow, Pytorch, Keras, CNTK

Academia : Web Platform for E-learning.

This paper illustrates how we can improve the existing manual system with the help of E-learning management system. The method aims to build an E-learning web application having better and safer user experience and provides an interactive teaching-learning platform for students and teachers. E-learning Management System is way of solving the educational problems using the modern technologies. It gives an error free, secure, reliable and fast management system. It can assist the user to concentrate on learning rather to concentrate on the record keeping and other stuff. It will help organization in better utilization of resources. Keywords: Web application, Database, backend, frontend, platform, E-learning, Frameworks

Professionals’ Digital Training for Child Maltreatment Prevention in the COVID-19 Era: A Pan-European Model

The responsiveness of professionals working with children and families is of key importance for child maltreatment early identification. However, this might be undermined when multifaceted circumstances, such as the COVID-19 pandemic, reduce interdisciplinary educational activities. Thanks to technological developments, digital platforms seem promising in dealing with new challenges for professionals’ training. We examined a digital approach to child maltreatment training through the ERICA project experience (Stopping Child Maltreatment through Pan-European Multiprofessional Training Programme). ERICA has been piloted during the pandemic in seven European centers involving interconnected sectors of professionals working with children and families. The training consisted of interactive modules embedded in a digital learning framework. Different aspects (technology, interaction, and organization) were evaluated and trainers’ feedback on digital features was sought. Technical issues were the main barrier, however, these did not significantly disrupt the training. The trainers perceived reduced interaction between participants, although distinct factors were uncovered as potential favorable mediators. Based on participants’ subjective experiences and perspectives, digital learning frameworks for professionals working with children and families (such as the ERICA model nested in its indispensable adaptation to an e-learning mode) can represent a novel interactive approach to empower trainers and trainees to tackle child maltreatment during critical times such as a pandemic, and as an alternative to more traditional learning frameworks.

Artificial Neural Networks and Deep Learning for Genomic Prediction of Continuous Outcomes

This chapter provides elements for implementing deep neural networks (deep learning) for continuous outcomes. We give details of the hyperparameters to be tuned in deep neural networks and provide a general guide for doing this task with more probability of success. Then we explain the most popular deep learning frameworks that can be used to implement these models as well as the most popular optimizers available in many software programs for deep learning. Several practical examples with plant breeding data for implementing deep neural networks in the Keras library are outlined. These examples take into account many components in the predictor as well many hyperparameters (hidden layer, number of neurons, learning rate, optimizers, penalization, etc.) for which we also illustrate how the tuning process can be done to increase the probability of a successful application.

Designing a Pathway to Digital Fluency

This chapter addresses the need for developing digital fluency skills in higher-education students in order to best prepare them for real-world success. The pathway to digital fluency is complex and requires a reimagined, collaborative approach to learning design. This chapter considers the elements of authentic learning as a means of developing students' digital fluency and proposes learning design as a pathway to action for teacher-developed authentic learning activities. The chapter begins by exploring the concept of digital fluency in order to develop a definition that informs pedagogical approach. Approaches to digital fluency development are examined through digital literacy and authentic learning frameworks. The pedagogical approach is further examined and conceptualized through the process of learning design. Considerations are presented at the end of each section to illustrate relationships between digital fluency, authentic learning, and learning design, and to allow for further exploration of concepts within unique contexts.