Browse Items (9795 total)
Sort by:
-
A Representation-Based Query Strategy to Derive Qualitative Features for Improved Churn Prediction
The effectiveness of any Machine Learning process depends on the accuracy of annotated data that is used to train a learner. However, manual annotation is expensive. Hence, researchers adopt a semi-supervised approach called active learning that aims to achieve state-of-the-art performance using minimal number of samples. Although it boosts classifier performance, the underlying query strategies are unable to eliminate redundancy in selected samples. Redundant samples lead to increased cost and sub-optimal performance of learner. Inspired by this challenge, the study proposes a new representation-based query strategy that selects highly informative and representative subsets of samples for manual annotation. Data comprises messages of a set of customers sent to a service provider. Series of experiments are conducted to analyze the effectiveness of the proposed query strategy, called 'Entropy-based Min Max Similarity' (E-MMSIM), in the context of topic classification for churn prediction. The foundation of E-MMSIM is an algorithm that is popularly used to sequence proteins in protein databases. The algorithm is modified and utilized to select the most representative and informative samples. The performance is evaluated using F1-score, AUC and accuracy. It is observed that 'E-MMSIM' outperforms popular query strategies, and improves performance of topic classifiers for each of the 4 topics of churn prediction. The trained topic classifiers are used to derive qualitative features. These features are further integrated with structured variables for the same group of customers to predict churn. Experiments provide evidence that inclusion of qualitative features derived using E-MMSIM, enhance the performance of churn classifiers by 5%. 2013 IEEE. -
Predicting customer churn: A systematic literature review
Churn prediction is an active topic for research and machine learning approaches have made significant contributions in this domain. Models built to address customer churn, aim to identify customers who are at a high risk of terminating services offered by a company. Hence, an effective machine learning model indirectly contributes to the revenue growth of an organization, by identifying at risk customers, well in advance. This improves the success rate of retention campaigns and reduces costs associated with churn. The aim of this study is to explore the state-of-the-art machine learning techniques used in churn prediction. A systematic literature review, that is driven by 5 research questions and rigorous quality assessment criteria, is presented. There are 38 primary studies that are selected out of 420 studies published between 2018 and 2021. The review identifies popular machine learning techniques used in churn prediction and provides directions for future research. Firstly, the study finds that churn models lack generalization capability across industry domains. Hence, it identifies a need for researchers to explore techniques that extend beyond model experimentation, to improve efficiency of classifiers across domains. Secondly, it is observed that the traditional approaches to churn prediction depend significantly on demographic, product-usage, and revenue features alone. However, recent papers have integrated social network analysis-related features in churn models and achieved satisfactory results. Furthermore, there is a lack of scientific work that utilizes information-rich content of customer-company-interaction instances via email, chat conversations and other means. This area is the least explored. Thirdly, there is scope to investigate the effect of hybrid sampling strategies on model performance. This has not been extensively evaluated in literature. Lastly, there is no formal guideline on correct evaluation parameters to be used for models applied on imbalanced churn datasets. This is a grey area that requires greater attention. 2022 Taru Publications. -
A Sampling-Based Stack Framework for Imbalanced Learning in Churn Prediction
Churn prediction is gaining popularity in the research community as a powerful paradigm that supports data-driven operational decisions. Datasets related to churn prediction are often skewed with imbalanced class distribution. Data-level solutions, like over-sampling and under-sampling, have been commonly used by researchers to address this problem. There are limited number of case studies that attempt to evolve these data-level solutions by integrating them with computationally advanced frameworks, like ensembles. Ensembles primarily employ algorithmic diversity using a fixed set of training instances to achieve superior performance. This study aims to introduce algorithmic diversity in ensembles by modifying the fixed set of training instances using diverse sampling strategies to increase predictive performance in imbalanced learning. Data is acquired from the world's largest open hotel commerce platform company. A four-part series of experiments is conducted to analyze the effectiveness of sampling techniques and ensemble solutions on model performance. A new sampling-based stack framework called 'Stacking of Samplers for Imbalanced Learning' is proposed. The framework combines the prediction capabilities of sampling solutions to stimulate the information gain of the meta features in ensemble. It is observed that the proposed framework leads to improvement in model performance with AUC of 86.4% and top-decile lift of 4.7 for customers of the hotel technology provider. Additionally, results show that the framework records a higher information gain for meta features used in a stack, compared to commonly used stack frameworks. 2013 IEEE. -
Effective ML Techniques to Predict Customer Churn
Customer churn is one of the most challenging problems that affects revenue and growth strategy of a company. According to a recent Gartner Tech Marketing survey, 91% of C-level respondents rate customer churn as one of their top concerns. However, only 43% have invested in additional resources to support customer expansion. Hence, retaining existing customers is of paramount importance to a company's growth. Many authors in the past have presented different versions of models to predict customer churn using machine learning techniques. The aim of this paper is to study some of the most important machine learning techniques used by researchers in the recent years. The paper also summarizes the prediction techniques, datasets used and performance achieved in these studies for a deeper understanding of the domain. The analysis shows that although hybrid and ensemble methods have been widely successful in improving model performance, there is a need for well-defined guidelines on appropriate model evaluation measures. While most approaches used are quantitative in nature, there is lack of research that focuses on information-rich content in customer company interaction instances, like emails, phone calls or customer support chat records. The information presented in the paper will not only help to increase awareness in industry about emerging trends in machine learning algorithms used in churn prediction, but also help new or existing researchers position their research activity appropriately. 2021 IEEE. -
Application of Machine Learning in Customer Churn Prediction
Retaining customers is the central component of a company's growth strategy. It is evident that several industries are experiencing a surge in customer churn due to the global pandemic. As a result, customer retention that lies at the core of customer relationship management, has become the foundation for every industry to plan for future growth. By reducing customer churn, a company can maximize its profit. Studies suggest that significant advancements are made in the field of customer churn prediction in domains like telecom, banking, e-commerce and energy sector. The focus of the paper is to present a detailed review of the various machine learning techniques applied to address churn. Fifty-five papers related to churn classification published between 2004 and 2020 are collected and analyzed. The reviewed papers are categorized into five main themes. These themes are feature selection techniques, methods to handle class imbalance, experimentation with machine learning algorithms, hybrid models and ensemble models respectively. Finally, few suggestions are presented as direction for future research. 2021 IEEE. -
Resilience in Children from Different Socioeconomic Backgrounds: An Exploratory Study
Poverty, violence, substance abuse, family dissonance and illness represent a few potential vulnerabilities in the lives of children who are at risk of failing in their future prospects. It is thus essential to explore resilience in children, owing to the excess or deficit of exposure and access in a childs life. This study aims at exploring the resilience of children of the age group 710years, from different socioeconomic backgrounds. The socioeconomic status was determined using the Kuppuswamy socioeconomic scale and these children had parents with authoritarian and permissive parenting styles which were screened through the Parenting Styles and Dimensions Questionnaire which act as risk factors for the children. Data was collected through individual semi-structured interviews with the participants and was analysed using thematic analysis. For the lower socioeconomic status group, the main themes identified were social interaction and competence, overcoming distress and future focus, and for the upper socioeconomic status group, the main themes identified were social interaction and competence and emotional management. The study paves the way for further exploration of the impact of economic status on childrens wellbeing and might inform changes at a clinical, research and policy level. 2023, The Author(s), under exclusive licence to Springer Nature Switzerland AG. -
Encoder-Decoder Approach toward Vehicle Detection
Vehicle Detection algorithms run on deep neural networks. But one problem arises, when the vehicle scale keeps on changing then we may get false detection or even sometimes no detection at all, especially when the object size is tiny. Then algorithms like CNN, fast-RCNN, and faster-RCNN have a high probability of missed detection. To tackle this situation YOLOv3 algorithm is being used. In the codec module, a multi-level feature pyramid is added to resolve multi-scale vehicle detection problems. The experiment was carried out with the KITTI dataset and it showed high accuracy in several environments including tiny vehicle objects. YOLOv3 was able to meet the application demand, especially in traffic surveillance Systems. Grenze Scientific Society, 2023. -
COVID-19 and stress of Indian youth: An association with background, on-line mode of teaching, resilience and hope
Background: COVID-19 pandemic causes serious threats to physical health and triggers wide varieties of psychological problems, including anxiety and depression. Youth exhibit a greater risk of developing psychological distress, especially during epidemics influencing their wellbeing. Objectives: To identify the relevant dimensions of psychological stress, mental health, hope and resilience and to examine the prevalence of stress in Indian youth and its relationship with socio-demographic information, online-mode of teaching, hope and resilience. Method: A cross-sectional online survey obtained information on socio-demographic background, online-mode of teaching, psychological stress, hope and resilience from the Indian youth. A Factor Analysis is also conducted on the recompenses of the Indian youth on psychological stress, mental health, hope and resilience separately to identify the major factors associated with parameters. The sample size in this study was 317, which is more than the required sample size (Tabachnik et al., 2001). Results: About 87% of the Indian youth perceived moderate to a high levels of psychological stress during the current COVID-19 pandemic. Different demographic, sociographic and psychographic segments were found to have high stress levels due to the pandemic, while psychological stress was found to be negatively correlated with resilience as well as hope. The findings identified significant dimensions of the stress caused by the pandemic and also identified the dimensions of mental health, resilience and hope among the study subjects. Conclusion: As stress has a long-term impact on human psychology and can disrupt the lives of people and as the findings suggest that the young population of the country have faced the greatest amount of stress during the pandemic, a greater need for mental health support is required to the young population, especially in post pandemic situations. The integration of online counselling and stress management programs could assist in mitigating the stress of youth involved in distance learning. 2023 The Author(s) -
COVID-19 and Mental Health of Indian Youth: Association with Background Variables and Stress
The coronavirus has become a public health concern of the decade, affecting the economic, social, and psychological stability of the whole world. Having understood the detrimental impact of the pandemic to the mental health of people of all age groups, youth is understood to be the most vulnerable population who receives its direct impact. The broad objective was to study the mental health status of Indian youth and its association with various demographic variables. Psychological stress and mental health was another relationship that was explored. A group of 317 participants between the age group of 19 to 29 voluntarily took part in the online survey. Gender was found to be associated with overall mental health status (p < 0.01) as well its correlates, namely anxiety (p < 0.05), depression (p < 0.05), and loss of behavioral control (p < 0.01). Association between age and loss of positive affect (p < 0.05), number of siblings and loss of behavioral control (p < 0.01), and family environment and overall mental health scores (p < 0.001) were found. Similarly, feeling of restlessness during lockdown (p < 0.001), availability of support (p < 0.001), and feeling the need to consult a mental health professional were associated with the overall mental health score as well as all its sub-scales. Further, there were strong negative correlations between psychological stress and overall mental health scores, as well as that of anxiety, depression, and loss of behavioral control and positive affect sub-scales. The study highlighted the need for psychological support services for the youth population of the country to cope with and adapt to the new situation. The Editor(s) (if applicable) and The Author(s), under exclusive license to Taylor and Francis Pte Ltd. 2022. -
SOCIAL PSYCHOLOGY: Theories and Applications
This book examines the concept of social psychology in today's context. It analyses the theoretical concepts of social psychology and their applicationto other fields. It further explores the discipline in a cultural, historical, and philosophical context with special emphasis on religion. The volume goes beyond individual focus and directs its attention to society as the centre of influence. It advocates for a symbiotic relationship between the concepts of social psychology and their implementation in a society transitioning from being value-oriented to commerce-oriented. The book also suggests ways in which social psychology can assist in dealing with issues plaguing today's world. This book will be useful to students of psychology, applied psychology, sociology, social work, public health, gender, and women studies. It will also be indispensable to professionals working in the field of paediatrics, forensic medicine, psychiatry, and law enforcement authorities like police and judiciary. 2024 Sibnath Deb, Anjali Gireesan, Pooja Prabhavalkar and Shayana Deb. -
A Cross-Sectional Study on Mental Health of School Students during the COVID-19 Pandemic in India
The broad objective of the present study is to assess the levels of anxiety and depression of school students during the COVID-19 lockdown phase and their association with students background, stress, concerns and social support. In this regard, the present study follows a novel two stage approach. In the first phase, an empirical survey was carried out, based on multivariate statistical analysis, wherein a group of 273 school students participated in the study voluntarily. In the second phase, a novel Picture Fuzzy FFA (PF-FFA) method was applied for understanding the dynamics of facilitating and prohibiting factors for three categories of focus groups (FG), formulated on the basis of attendance in online classes. Findings revealed a significant impact of anxiety and depression on mental health. Further, PF-FFA examinedthe impact of the driving forces that steered children to attend class as contrasted to the the impact of the restricting forces. 2022 by the authors. -
Cloud computing security for public cloud using ciphers and queueing petri nets
Cloud computing is the most used word in the domain of Information Technology, which is making colossal differentiations in the IT business. Nowadays, a massive proportion of data is being made, and the masters are discovering better approaches for managing this data. In a general sense, the word cloud implies a virtual database that stores immense data from various clients. There are three sorts of cloud public, private and hybrid. A public cloud is fundamental for general customers where customers can use cloud benefits free or by paying. Private cloud is for explicit associations, and hybrid one is in a broad sense a mix of both. Cloud offers diverse kind of administrations, for instance, IAAS, PAAS, SAAS where administrations like a stage for running any application, getting to the enormous information extra room, can use any application running under the cloud are given. The cloud similarly has a shortcoming concerning the security for the data warehouse. In a general sense, public cloud is inclined to data modification, data hacking and therefore, the integrity and privacy of the data are being undermined. Here in our work our motive is to verify the information that will be taken care of in the public cloud by using the multi-stage encryption. The estimation that we have proposed is a mix of Rail Fence cipher and Play Fair cipher. 2020, IJSTR. -
School corporal punishment, family tension, and students internalizing problems: Evidence from India
There is considerable evidence that parental corporal punishment (CP) is positively associated with childrens behavioral and mental health problems. However, there is very little evidence addressing whether CP perpetrated by teachers or school staff is similarly associated with problematic student functioning. To address this gap in the research literature, data were collected from students in a locale where school CP continues to be widely practiced. Participants were 519 adolescents attending public or private schools in Puducherry, a city in eastern India. Students completed surveys assessing school CP, internalizing problems, social support, and resilience. The results indicated that 62% of the students reported experiencing school CP in the past 12 months, with males and those attending public schools being significantly more likely to report school CP than females and those in private schools. Youth who reported school CP reported more anxiety and depression. That relation was more pronounced in youth who reported family tension. Social support and resilience did not moderate the relations. The findings add to the substantial evidence about negative associations regarding the use of CP but in a new venuethe school, and provide some evidence for the need to change how students are disciplined in schools in India and elsewhere. 2016, The Author(s) 2016. -
Youth of North East States of India: Issues, Concerns and Need for Mental Health Support as Perceived by NCC Officers
The youth of North-East India are in disadvantaged situations as compared to youth from the rest of the country in all respects. The objective of this article was to examine the views of the NCC Officers of North-East states about youth welfare in the region as they have first-hand experience in dealing with youth. Participants views were obtained on-line, by using a Semi-structured Questionnaire in the form of Google Form. A group of 142 NCC Officers provided feedback. Data collected were subjected to thematic analysis. Findings disclosed that youth of North-East states experience a range of challenges including poverty, lack of internet facilities, inability to attend NCC camps due to ongoing classes, substance dependence, lack of guidance and support leading to dropout and lack of values. The NCC Officers opined that a good number of North-East youths require mental health support and career guidance, in addition to mental health awareness. 2023 Taylor & Francis Group, LLC. -
Doctoral Research by Youth: Analyzing the Role of Socio-Demographic Variables on Flourishing and Grit
The study examines the importance of socio-demographic variables like age, gender, family environment, and relationship with parents and friends in deter-mining non-cognitive traits such as flourishing and grit, during the tenure of doctoral research. The cross-sectional correlational study comprises 400 Ph.D. scholars from a Central University in India, who were given a personal data sheet, the Flourishing Scale and the Grit Scale, for assessment. The results of the F-test showed that flourishing was significantly related to age, family environment and relationship with friends, and grit was significantly related to family environment and relationship with friends. Analysis using Pearson correlation found a weak correlation between flourishing and the three subscales of grit, namely ambition, consistency of interest, and perseverance of effort. Findings suggest that the socio-demographic variables are important contributors in the long-term goal-oriented behaviors and that flourishing and grit are two related but not correlated variables that influence completion and attrition of the doctoral research. The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023. -
Happiness, Meaning, and Satisfaction in Life as Perceived by Indian University Students and Their Association with Spirituality
The present study aims to examine the association between various dimensions of psychological well-being (subjective happiness, satisfaction, and meaning in life), spirituality, and demographic and socioeconomic background of university students. A total of 414 postgraduate students were selected from three different schools, viz. science, management, and social sciences/humanities of Pondicherry University (A Central University), Puducherry, India, following multistage cluster sampling method. One semi-structured questionnaire and four standardized psychological scales, viz. subjective happiness scale, satisfaction with life scale, meaning in life questionnaire, and spirituality attitude inventory, were used for data collection after checking psychometric properties of the scales. The results show that a positive significant correlation between spirituality and subjective happiness exists. Spirituality is also correlated with meaning in life and satisfaction with life scale. Statistically, no significant gender difference was observed with respect to subjective happiness, meaning, and satisfaction in life as well as spirituality although the mean score of female students was more in all the four psychological domains. Non-integrated students are found to be happier than integrated students, and statistically it was significant. Positive interpersonal relationship and congenial family environment were probed to be facilitating factors for positive mental health of university students. There is a severe need to address students mental health by every educational institution through multiple programs. 2019, Springer Science+Business Media, LLC, part of Springer Nature. -
Level of green computing based management practices for digital revolution and New India /
International Journal of engineerig And Advanced Technology, Vol.8, Issue 3, pp.133-136, ISSN No: 2249-8958. -
Facial Expression Recognition with Transfer Learning: A Deep Dive
In the realm of affective computing, where the nuanced interpretation of facial expressions plays a pivotal role, this research presents a comprehensive methodology aimed at refining the precision of facial expression recognition on the CK+ (Cohn-Kanade Extended) dataset. Our method uses the robust DenseNet121 architecture that has been pretrained on the 'imagenet' dataset, and it leverages transfer learning on the foundational CK+ dataset. The model deftly handles issues with overfitting, normalization, and feature extraction that are present in facial expression detection on CK+. Our approach not only achieves an overall accuracy of 98%, with a 5.86% accuracy enhancement over the base paper on the CK+ dataset, but also shows remarkable precision, recall, and F1-score values for individual emotion classes. It is noteworthy that emotions such as anger, contempt, and disgust have precision rates that reach 100%. The study highlights the model's noteworthy input to affective computing and presents its possible real-world uses in emotion analysis on CK+ and human-computer interaction. 2024 IEEE. -
Pandemic, theatre and performance: Democratizing the subalterns through the Theatre of the Oppressed
The presented work analyses Theatre of the Oppressed (TO) methods impacting the pandemic. It follows the WHO timeline, when the COVID-19 pandemic had cast a dark shadow, making sustenance difficult for the marginalized section of Indian society. TO methods, though reflected, adapted and accommodated exhaustively in Indian applied theatre over the last four decades, offered a fresh, collective, democratic space during the pandemic. Forum theatre (FT) and legislative theatre (LT) praxis rendered a platform for activism, awareness and emancipation of the subalterns during the pandemic. Thus, TO renewed psycho-social dialogue and critical, creative, experimental space during this time. The applicability of such methods facilitating social change is gauged using Boals spect-actorship and Freires conscientization. The article looks forward to the TO signposts to serve as nodal points for further scholarly discussion and study on democratizing the disenfranchised population through FT and LT during the pandemic. 2023 Intellect Ltd Article. English language. All Rights Reserved. -
Sentiment Analysis for Online Shopping Reviews Using Machine Learning
Everyday shoppers need reliable and insightful reviews of e-commerce websites to enhance their shopping experience. This research study explores sentiment analysis on Amazon reviews. It utilizes them as a diverse repository of customer opinions by unlocking their embedded sentiments, thereby recognizing their pivotal role in guiding potential buyers. Sentiment misinterpretations may result from many machine learning models that have trouble comprehending the context of Amazon reviews, particularly regarding subtle wordings, sarcasm, or irony. Additionally, these models can have biases that skew sentiment analysis results, mainly when working with a diverse set of Amazon review datasets. To overcome these, three machine learning models, namely, Bidirectional Encoder Representations from Transformers (BERT), Bidirectional and Auto-Regressive Transformers (BART), and Generative Pre-trained Transformers (GPT) are used in this study. During the experimental research, it was observed that BERT gave the highest accuracy of 90% when compared with BART (70%) and GPT (84%) models. BERTs bidirectional contextual comprehension at identifying subtleties in language provides a thorough and realistic representation of the sentiments of Amazon users, which is why the model gave the highest accuracy. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.