Original Article
Masoumeh Moterased; Seyed Mojtaba Sajadi; Ali Davari; Mohammad Reza Zali
Abstract
This study discusses the prediction model of Entrepreneurial Exit from Entrepreneurial Perceptions, acquired the data from the Global Entrepreneurship Monitor's (GEM) database in 2008-2019. Some essential indicators include Opportunity Perception, Fear of Failure, Capability Perception, Role Model, and ...
Read More
This study discusses the prediction model of Entrepreneurial Exit from Entrepreneurial Perceptions, acquired the data from the Global Entrepreneurship Monitor's (GEM) database in 2008-2019. Some essential indicators include Opportunity Perception, Fear of Failure, Capability Perception, Role Model, and Entrepreneurial Intention. Data mining results show that the exit reasons and entrepreneurial intention have a more significant impact on entrepreneurial exit than other variables. This research applies the Random Forest Algorithm to get a prediction model that shows the entrepreneurial exit. According to the Random Forest Algorithm results, accuracy, ROC-AUC score, AUC curve, precision, recall, and F1 score validate the classification method. The prediction model shows that the best accuracy predictor of entrepreneurial exit is 99 percent, and another criteria ROC_AUC score 96%. Consistent results demonstrate that the proposed method can consider a promisingly successful predictive model of entrepreneurial exit with excellent predictive performance. These results can predict the individuals' entrepreneurial exit possibility before the psychological and financial impact and loss of capital and failure.
Original Article
Fatemeh Mohades Deilami; Hossein Sadr; Mozhdeh Nazari
Abstract
Personality can be defined as the combination of behavior, emotion, motivation, and thoughts that aim at describing various aspects of human behavior based on a few stable and measurable characteristics. Considering the fact that our personality has a remarkable influence in our daily life, automatic ...
Read More
Personality can be defined as the combination of behavior, emotion, motivation, and thoughts that aim at describing various aspects of human behavior based on a few stable and measurable characteristics. Considering the fact that our personality has a remarkable influence in our daily life, automatic recognition of a person's personality attributes can provide many essential practical applications in various aspects of cognitive science. Although various methods have been recently proposed for the task of personality recognition, most of them have mainly focused on human-designed statistical features and they did not make use of rich semantic information existing in users' generated texts while not only these contents can demonstrate its writer's internal thought and emotion but also can be assumed as the most direct way for people to state their feeling and opinion in an understandable form. In order to make use of this valuable semantic information as well as overcoming the complexity and handcraft feature requirement of previous methods, a deep learning based method for the task of personality recognition from text is proposed in this paper. Among various deep neural networks, Convolutional Neural Networks (CNN) have demonstrated profound efficiency in natural language processing and especially personality detection. Owing to the fact that various filter sizes in CNN may influence its performance, we decided to combine CNN with AdaBoost, a classical ensemble algorithm, to consider the possibility of using the contribution of various filter lengths and gasp their potential in the final classification via combining various classifiers with respective filter size using AdaBoost. Our proposed method was validated on the Essay dataset by conducting a series of experiments and the empirical results demonstrated the superiority of our proposed method compared to both machine learning and deep learning methods for the task of personality recognition.
Original Article
Agyan Panda; Bharath Yadlapalli; Zhi Zhou
Abstract
Every year, millions of dollars are lost due to fraudulent credit card transactions. To help fraud investigators, more algorithms are turning to powerful machine learning methodologies. Designing fraud detection algorithms is particularly difficult because to the non-stationary distribution of data, ...
Read More
Every year, millions of dollars are lost due to fraudulent credit card transactions. To help fraud investigators, more algorithms are turning to powerful machine learning methodologies. Designing fraud detection algorithms is particularly difficult because to the non-stationary distribution of data, excessively skewed class distributions, and continuous streams of transactions. At the same time, due to confidentiality considerations, public data is uncommon, leaving many questions unanswered about the best technique for dealing with them. We present some replies from the practitioners in this publication. Un balanced ness, non- stationarity and assessment. Our industrial partner provided us with an actual credit card dataset, which we used to do the analysis. In this project, we attempt to develop and evaluate a model for the imbalanced credit card fraud dataset.
Original Article
Mohammad Reza Mozaffari; Sahar Ostovan
Abstract
Data envelopment analysis based on mathematical programming for decision-making units determines the efficiency score in addition to the projection of inefficient DMUs on the efficient frontier. Centralized Allocation Resource (CRA) with a two-stage linear programming model captures the projection of ...
Read More
Data envelopment analysis based on mathematical programming for decision-making units determines the efficiency score in addition to the projection of inefficient DMUs on the efficient frontier. Centralized Allocation Resource (CRA) with a two-stage linear programming model captures the projection of DMUs on the efficient frontier. But since the input and output vectors of each DMU in the DEA are crucial, they may be random data that follow a particular distribution. Hence, many applied studies face random data. This paper shows a two-stage supply chain with random data and the CRA model with ratio data has been used to calculate the projection of DMUs. In the end, the supply chain of 11 Iranian Airlines with random data during the period of 2011-2017 was considered concerning sustainability factors.
Original Article
Roohollah Abbasi Shureshjani; Bita Shakouri
Abstract
This note shows that the parametric ranking method proposed by Darehmiraki [M. Darehmiraki, A novel parametric ranking method for intuitionistic fuzzy numbers, Iranian Journal of Fuzzy Systems, 16(1) (2019), 129-143] is not correct. By two appropriate examples, we show that the developed index for ranking ...
Read More
This note shows that the parametric ranking method proposed by Darehmiraki [M. Darehmiraki, A novel parametric ranking method for intuitionistic fuzzy numbers, Iranian Journal of Fuzzy Systems, 16(1) (2019), 129-143] is not correct. By two appropriate examples, we show that the developed index for ranking intuitionistic fuzzy numbers is not suitable and implies wrong results.
Original Article
Fatemeh Taghvaei; Ramin Safa
Abstract
As the construction sector accounts for the highest energy consumption worldwide, new solutions must be offered in buildings through the adoption of energy-efficient techniques. The main factors involved in energy consumption and residents' behaviors patterns considering environmentally-friendly lifestyle ...
Read More
As the construction sector accounts for the highest energy consumption worldwide, new solutions must be offered in buildings through the adoption of energy-efficient techniques. The main factors involved in energy consumption and residents' behaviors patterns considering environmentally-friendly lifestyle changes must be clearly identified and modeled to provide such solutions. One of the most important topics in smart grids is managing energy consumption in buildings, and one way to optimize energy consumption by analyzing building energy data is to use personalized recommender systems. The Non-Intrusive Load Monitoring (NILM) technique is an important way to cost-effective real-time monitoring the energy consumption and time of use for each appliance. However, the combination of recommender systems and NILM has received less attention. In this paper, a personalized NILM-based recommender system is proposed, which has three main phases: DAE-based NILM, TF-IDF-based text classification, and personalized recommender system. The proposed approach is investigated using the Reference Energy Disaggregation Dataset (REDD). According to the results, the accuracy of the proposed framework is about 60%.