Journal of Information and Organizational Sciences

293-304

Authors: Paweł Baranowski, Adrian Stepniak

Deep convolutional neural networks (CNNs) became an industry standard in image processing. However, in order to keep their high efficiency, a large annotated sample is required in the case of supervised learning. In this paper we apply the techniques specific for relatively small sample to a court files dataset. Specifically, we propose transfer learning and semisupervised learning to classify scanned page as having a table or not. We use four CNNs architectures established in the literature and find that transfer learning improves the classification performance, compared to the fully supervised learning. This result is especially evident in the scenarios where only a part of convolutioanl layers are transferred. The gains from semisupervised learning are ambiguous, as the results vary over CNNs architectures. Overall, our results show that office documents classification can achieve high accuracy when transferring initial convolutional layers is applied.

323-344

Authors: Aakansha Singh, Harminder Kaur Gujral

The present research aims to study the relation between emotional intelligence and organizational culture during the Covid-19 pandemic. Emotional intelligence is an individual's ability to understand and manage their own and others' emotions. This ability can be instrumental in driving the culture of universities and colleges. During this crucial time of Covid-19, there is a sudden shift from traditional/ face-to-face teaching to online/hybrid education; it is vital to provide a congenial environment and culture to teachers and students to promote a smooth and amiable teaching/learning experience. The data was collected from a sample of 10 colleges of higher education from NCR. The respondents included academicians of different academic ranks. 232 questionnaires were received, out of which 213 respondent's data was usable. Convenience sampling was used in this research to collect data from the colleges. According to the findings, there is a strong relation between emotional intelligence and organizational culture. The organizational culture and its dimensions are strongly affected by emotional intelligence. The most robust relationship exists between managing emotions and managing change in the organization.

361-375

Authors: Blaženka Divjak, Darko Grabar, Barbi Svetec, Petra Vondra

We present a comprehensive learning design (LD) concept and tool, motivated by the needs identified by higher education (HE) practitioners. The concept and tool aim at implementing contemporary research findings and theory to support balanced LD planning (BDP). The student-centered BDP concept and tool provide innovation to LD planning by strongly focusing on learning outcomes (LOs) and student workload, aligning study program and course level LOs, ensuring constructive alignment and assessment validity, enhancing LD by using learning analytics, and enabling flexible use in different contexts and pedagogical approaches. The ongoing work has been done according to design science methodology, with positive first feedback from HE practitioners. We identify areas for further research and improvement, including testing the BDP tool in real-world HE contexts and its integration with learning management systems. This could help close the gap between intended (often innovative) LDs and their implementation in real teching and learning environments.

391-420

Authors: Eda Sultan Arar, Serap Şahin

The aim of the research is to examine the work ethics perceptions of academics from different generations and the relationship between these perceptions and workplace deviation behaviors. In line with the purpose, the questionnaire prepared by using 5 questions with demographic variables, academic ethical values scale and workplace deviation behavior scale were applied to 472 academicians working at state universities in Ankara and the nearby provinces. The results revealed moderately negative relationship between academic ethical values and one of its sub-dimensions which is the values for the institution and workplace deviation behaviors. Moreover, weak and strong negative relationships were found between academic ethical values and other sub-factors of workplace deviation behaviors. Intergenerational differences were found between academic ethical values and academic ethical values towards the teaching process and serious workplace deviation towards the organization, and between academic ethical values towards colleagues and deviant behaviors towards the organization.

433-451

Authors: Manish Mohan, Anil Jadhav

No industry can thrive without customers and with customers comes the chances of customer churn. Since customer churn have direct-impact on the revenue, all the industries are focusing in understanding the factors influencing churn and are developing methods to predict the customer churn effectively. Today, never as before, customers have wide variety of options to choose between any service or product. In addition, nowadays customers enjoy multiple subscriptions of service providers across sectors. In this study we aim to identify: i) Factors influencing customer churn on OTT platform, and ii) Predict customer churn on OTT platform. The data for this study is collected from 317 respondents, using questionnaire method, who have multiple OTT platform subscription. The questionnaire data contains 19 items which includes demographic features, usage of OTG platform, and user contentment factors about OTT service. We have identified factors influencing customer churn in Over-The-Top (OTT) platform by combining Recursive Feature Elimination (RFE), Linear Regression, and Ridge Regression feature ranking methods. We have used Hierarchical Logistic Regression, to understand impact of two newly introduced factors namely 'Multiple Subscription' and 'Switching Frequency' on the overall performance of the customer churn prediction. Finally, customer churn prediction is done using Decision Tree, Random Forest, AdaBoost, and Gradient boosting techniques. We found that random forest method gives better prediction results.

473-489

Authors: Farouk Ouatik, Mohammed Erritali, Fahd Ouatik, Mostafa Jourhmane

The process of traditional school guidance is carried out by specialists. They make a study of the student files based on the marks of the first year and the second year of the baccalaureate. Considering the progressive number of students and also the lack of time to make the decision, as the selection of the specialty has a great effect on the academic course of the students, we have realized a system of specialty recommendation, to computerize the orientation process and save time. But the major problem is that the students do not care about this process and pay no attention to it despite its importance. As well as the software which makes the orientation is chargeable. To order these specialties to take the best specialty. We have arrived at a problem of multi-criteria which makes it impossible to make a decision with these criteria. Because these criteria do not have the same importance and also are not compatible, as there are criteria that must be maximum and other criteria must be minimal. To solve this problem, two systems of orientation and academic reorientation of students have been implemented. In both systems, the SMOTE method has been used to balance the learning data in the preprocessing phase. Then in the treatment phase, we sorted the specialties using in the first system, a hybridization of TOPSIS method and the information gain to find the weights of the criteria used, and in the second system, we used a hybridization of the AHP method and information gain. The results obtained indicate that before balancing data using the SMOTE method, the total accuracy of TOPSIS (84.20%) is higher than the total accuracy of AHP (83.71%). After applying balancing data using the SMOTE method, the total accuracy has increased. The total accuracy of TOPSIS (91.35%) is also higher than the total accuracy of AHP (90.83%). For the complexity of the two methods, it is related to the number of criteria and the number of alternatives. If the number of criteria is more than 10 criteria, the complexity of TOPSIS is less than the complexity of AHP, and vice versa. The complexity of the two methods also depends on the number of alternatives, if the number of alternatives exceeds 10, the TOPSIS method becomes more complex than the AHP method. In general, the system based on TOPSIS method and the information gain is more precise than the system based on the AHP method and the information gain. But the complexity of the AHP method is less than the complexity of the TOPSIS method.