Search results

With the advancements in photo editing software, it is possible to generate fake images, degrading the trust in digital images. Forged images, which appear like authentic images, can be created without leaving any visual clues about the alteration in the image. Image forensic field has introduced several forgery detection techniques, which effectively distinguish fake images from the original ones, to restore the trust in digital images. Among several forgery images, spliced images involving human faces are more unsafe. Hence, there is a need for a forgery detection approach to detect the spliced images.

This paper proposes a Taylor-rider optimization algorithm-based deep convolutional neural network (Taylor-ROA-based DeepCNN) for detecting spliced images. Initially, the human faces in the spliced images are detected using the Viola–Jones algorithm, from which the 3-dimensional (3D) shape of the face is established using landmark-based 3D morphable model (L3DMM), which estimates the light coefficients. Then, the distance measures, such as Bhattacharya, Seuclidean, Euclidean, Hamming, Chebyshev and correlation coefficients are determined from the light coefficients of the faces. These form the feature vector to the proposed Taylor-ROA-based DeepCNN, which determines the spliced images.

Experimental analysis using DSO-1, DSI-1, real dataset and hybrid dataset reveal that the proposed approach acquired the maximal accuracy, true positive rate (TPR) and true negative rate (TNR) of 99%, 98.88% and 96.03%, respectively, for DSO-1 dataset. The proposed method reached the performance improvement of 24.49%, 8.92%, 6.72%, 4.17%, 0.25%, 0.13%, 0.06%, and 0.06% in comparison to the existing methods, such as Kee and Farid's, shape from shading (SFS), random guess, Bo Peng et al., neural network, FOA-SVNN, CNN-based MBK, and Manoj Kumar et al., respectively, in terms of accuracy.

The Taylor-ROA is developed by integrating the Taylor series in rider optimization algorithm (ROA) for optimally tuning the DeepCNN.

The purpose of this paper is to conceptualize the influencing characteristics of user-generated content over perceived structure of social platforms to plan various business practices to improve the social commerce activities especially in emerging markets.

In first step, key factors are identified from the systematic literature studies and experts’ opinion. Second, total interpretive structural modeling is used to interpret the complexity of relationships among various factors. Further, fuzzy-MICMAC analysis is used to determine the most driving factors to improve the social commerce.

The study illustrates hierarchical relationships that can help policy designers, business leaders and decision makers to gain maximum benefits of mining the user-generated content to implement the people-oriented changes to improve economic capability of a country.

The applicability of the study findings can be deployed mostly in emerging economies. The study has not considered the data security, cyberbullying, authenticity and reliability of shared content spread virally over social sites.

The paper will assist policy makers, business planners and solution providers to give importance to public-led influencing patterns to offer business solutions based on people demand and choices.

The customers’ engagement, sharing reviews, getting suggestions from other people and finding new trends through user-generated content would enable higher social commerce activities.

The paper develops hierarchical relationships to represent the advancements in business and marketing activities based on influencing patterns and behavior of users over social sites.

Soy beverage is becoming more and more popular because it is touted as a healthy food containing useful phytochemicals and is free from lactose and cholesterol. The purpose of this paper is to optimize the spray drying process parameters for obtaining soy beverage powder with good reconstitution and handling properties.

Pre-concentrated soy beverage was dried in a laboratory model spray dryer, and the effects of inlet air temperature (180-220°C), feed rate (20-40 ml/min) and feed solid content (15-25 per cent) on some physical parameters and reconstitution properties (wettability and dispersibility) of spray-dried soy beverage powders were investigated. Second order polynomial response surface model was selected for the analysis of data and optimization of the process.

Spray drying of soy beverage at different processing conditions resulted in powders with particle size (volume mean diameter) in the range of 86 to 156 µm. Dispersibility and wetting time of the spray-dried soy beverage powders was found to be in the range of 56 to 78 per cent and 30 to 90 s respectively, under various drying conditions. Inlet air temperature was found to be the main factor affecting most of the quality parameters, followed by solid content of the feed. Temperature significantly affected the wettability, dispersibility, colour parameters, particle size and flowability of the powder at p ≤ 0.01. Lower temperature and higher feed solid content produced bigger-sized powder particles with better handling properties in terms of flowability and cohesiveness. A moderate inlet air temperature (196°C), higher feed solid content (24 per cent) and lower feed rate (27 ml/min) were found suitable for drying of soy beverage.

The study implied the possibility of producing powder from soy beverage using the spray-drying method and optimized drying conditions for obtaining soy beverage powder with good reconstitution properties.

The finding of this study demonstrated for the first time how the inlet air temperature, feed solid content and feed rate during spray-drying influenced different quality parameters of soy beverage powder. Further, an optimized drying condition has been identified.

Artificial intelligence (AI) is increasingly applied by businesses to optimize their processes and decision-making, develop effective and efficient strategies, and positively influence customer behaviors. Businesses use AI to generate behaviors such as customer retention (CR). The existing literature on “AI and CR” is vastly scattered. The paper aims to review the present research on AI in CR systematically and suggest future research directions to further develop the field.

The Scopus database is used to collect the data for systematic review and bibliometric analysis using the VOSviewer tool. The paper performs the following analysis: (1) year-wise publications and citations, (2) co-authorship analysis of authors, countries, and affiliations, (3) citation analysis of articles and journals, (4) co-occurrence visualization of binding terms, and (5) bibliographic coupling of articles.

Five research themes are identified, namely, (1) AI and customer churn prediction in CR, (2) AI and customer service experience in CR, (3) AI and customer sentiment analysis in CR, (4) AI and customer (big data) analytics in CR, and (5) AI privacy and ethical concerns in CR. Based on the research themes, fifteen future research objectives and a future research framework are suggested.

The paper has important implications for researchers and managers as it reveals vital insights into the latest trends and paths in AI-CR research and practices. It focuses on privacy and ethical issues of AI; hence, it will help the government develop policies for sustainable AI adoption for CR.

To the author's best knowledge, this paper is the first attempt to comprehensively review the existing research on “AI and CR” using bibliometric analysis.

Vaccine hesitancy is a long-standing issue among both the general population and health communicators. This study aims to ascertain the inclination and the reasons for vaccine hesitancy by conducting content analysis and sentiment analysis of the perspectives expressed in comments on videos related to vaccine hesitancy uploaded from India on YouTube.

The assessment of the sentiments of the vaccine-hesitant population is done using Valence Aware Dictionary and sEntiment Reasoner sentiment analysis module implemented with Python’s NLTK library to automatically determine the sentiments of the comments. Manual content analysis was performed on 60.09% viewer comments randomly selected from the total comments in 238 videos on vaccine hesitancy originated from India and labelled each comment with labels “Anti”, “Pro”, “Confused”, “Not Applicable” and “Unrelated” labels.

The study found “Mistrust-Government policies”, “Fear-health related consequences”, “Mistrust-Scientific research”, “Vaccine effectiveness and efficacy” and “Misinformation/myths” as the top five determinants for vaccine hesitancy, whereas “Religious beliefs”, “Fear-Economic consequences”, “Side Effects- short-term” and “Fear-mode of administration” found to be the lesser cited reasons for vaccine hesitancy. However, the study also investigates changes in the inclination of Indian commenters towards vaccine hesitancy and revolving issues over time.

Public health policymakers and health communicators may find the study useful in determining vaccine hesitancy factors in India.

The originality of this study lies in its approach. To date, no sentiment analysis has been conducted on the content released on YouTube by Indian content creators regarding pro- and anti-vaccination videos. This inquiry seeks to fill this research gap.

The goal of this research is to look at how urban microfinance affects livelihood transformation in terms of poverty reduction, living standards, social well-being, empowerment and entrepreneurship.

This paper analyses the role of urban microfinance towards livelihood with special reference to Western Uttar Pradesh. Primary data were collected from 321 respondents who are users of a microfinance programme using a standardised questionnaire. The data were collected using a stratified random sampling technique, and the data were analysed using structural equation modelling.

Urban microfinance has a considerable impact on poverty reduction, the standard of living, social well-being, empowerment and entrepreneurship in the urban poor, according to the findings.

The fact that the majority of the borrowers were uneducated was the most significant barrier to them filling out the questionnaire. Their anxiety was the most significant psychological obstacle to successfully answering the questions, and it took time. As a result, it is urged that proper counselling be conducted before the poor borrowers fill out the questionnaire.

The current study highlights the factors that lead to the utilisation of microfinance services. This research will aid MFIs in selecting the appropriate products and services for the urban poor. The results of this study will aid them in understanding and meeting the expectations of microfinance CEOs.

This is a first study conducted in Northern zone of India measuring the roles urban microfinance institutions (MFIs) in uplifting the livelihood of urban poor.

This paper aims to identify the key factors to design efficient, healthy and potentially economical neighbourhood places in the surroundings of smart cities to reduce the urban polarization for the sustainable urban development.

A two-stage methodology is followed. First, the key factors for neighbourhood are identified from literature studies. The selected factors are validated by sample t-tests. Second, the total interpretive structural modeling is used to interpret the complexity of relationships among various factors. Further, cross-impact matrix multiplication is applied for classification analysis to find the most driving factors for neighbourhood design.

The contribution of this research is to show hierarchical relationships among the various factors to design the neighbourhood places as smart from the perspectives of city planners and decision makers.

The applicability of the research findings is limited to developing countries mainly where population is large and most of cities have high pressure on its infrastructure to fulfil the citizens’ demands.

This paper will aid policymakers, city planners and government officials to design a sustainable smart city model in which smart neighbourhood would also be the potential solution to decrease pressure on a city’s critical infrastructure especially in developing countries.

A smart city could be considered as the centre point of smart initiatives to develop a place smart, and it should continue beyond the city boundaries to enhance the facilities, services, resources utilization and working environment in neighbourhood places also.

The study explores the various literature on neighbourhood planning and then link with smart city development as current need of urban development scenario. The authors propose a hierarchical relation framework to develop the neighbourhood places as smart places to meet the future demand of urbanization in developing countries like India.

This bibliometric study aims to analyze publication trends, active countries, collaborations, influential citations and thematic evolution in learning analytics (LA) research focused on higher education (HE) during and after the COVID-19 lockdown period.

From the Scopus database, this bibliometric analysis extracts and evaluates 609 scholarly articles on LA in HE starting in 2019. The multidimensional process identifies the scope impacts, advancing the understanding of LA in HE. An analysis of co-citation data uncovers the key influences that have shaped the literature. This study uses the stimulus-organism-response (SOR) theory to suggest future research directions and organizational adaptations to new LA technologies and learner responses to LA-enabled personalized interventions.

Learning analytics are becoming important in the HE environment during and after the COVID-19 lockout. Institutions have used LA to collect socio-technical data from digital platforms, giving them important insights into learning processes and systems. The data gathered through LA has assisted in identifying areas for development, opening the path for improved student success and academic performance evaluation and helping students transition to the workforce.

The study’s concentration on the post-COVID-19 timeframe may lead to paying attention to potential pandemic developments. Nonetheless, the findings provide a thorough picture of LA’s contributions to HE and valuable ideas for future study initiatives. Future research with the SOR framework suggests areas for additional study to maximize LA’s potential in diverse HE situations.

This study adds to the growing corpus of knowledge on learning analytics in HE, especially in light of the COVID-19 lockdown and its aftermath. By using bibliometric analysis, the study provides a complete and evidence-based understanding of how LA has been used to address challenges related to HE. This study uses bibliometric analysis and SOR theory to appraise and map HE learning analytics research. The selected study themes can help scholars, educators and institutions shape their future efforts to improve teaching, learning and support mechanisms through learning analytics.

The purpose of this study is to review the literature and practice of upcycling. In particular, the objective of this study is threefold: to comprehend the concept of upcycling and, subsequently, understanding the prominent terminologies used in the literature; to understand the process of upcycling and problem associated with it; and to review current literature and practice of upcycling for clothes.

A scientific literature review procedure proposed by Mayring (2002) was adopted to select and screen the paper that comprises the following steps: material collection, descriptive analysis and material evaluation.

Upcycling literature has witnessed significant contribution in the past one decade. The paper has identified various terminologies and definitions such as recycling, down-cycling, upcycling and redesign, which are used in the literature.

The present study may help the scholars to understand the current state of literature. A practitioner of upcycling can use the findings to improve and standardise the existing process.

The process of redesigning is one of the important steps in upcycling, which comprises ideation, reconstruction and fitting. The limitation of redesigning is variability in size and pattern. This can be overcome through various techniques such as craftsmanship, time, innovation, provenance, desire and narrative.

The purpose of this paper is to examine the role of power on supply chain performance (SCP) in the context of small and medium sized agribusiness enterprises (SMEs). Contrary to most of previous studies, which collect and analyze data from one side of a relationship dyad using a focal firm approach, a matched triad approach was employed in data collection and analysis.

Empirical data was collected from 150 agribusiness supply chain members from the maize supply chain in Uganda. Analysis was done using multi-group analysis and structural equations modeling.

Results highlight the differences in the perception of power use and how it influences SCP. The differences in perception suggest the existence of power asymmetry amongst supply chain members. This work contributes to the ongoing debate concerning the use of triad as a unit of analysis as opposed to a firm or a dyad.

This study only focused on one commodity chain in one country, which can limit the broad application of the findings.

The novelty of this work lies in fact that the authors assess perception of power amongst supply chain members in a triadic context, a perspective that has not been adequately tested in agribusiness supply chain management studies before.