m. j. tarokh

Customer churn prediction has been gaining significant attention due to the increasing competition among mobile service providers. Machine learning algorithms are commonly used to predict churn; however, their performance can still be improved due to the complexity of customer data structure. Additionally, the lack of interpretability in their results leads to a lack of trust among managers. In this study, a step-by-step framework consisting of three layers is proposed to predict customer churn with high interpretability. The first layer utilizes data preprocessing techniques, the second layer proposes a novel classification model based on supervised and unsupervised algorithms, and the third layer uses evaluation criteria to improve interpretability. The proposed model outperforms existing models in both predictive and descriptive scores. The novelties of this paper lie in proposing a hybrid machine learning model for customer churn prediction and evaluating its interpretability using extracted indicators. Results demonstrate the superiority of clustered dataset versions of models over non-clustered versions, with KNN achieving a recall score of almost 99% for the first layer and the cluster decision tree achieving a 96% recall score for the second layer. Additionally, parameter sensitivity and stability are found to be effective interpretability evaluation metrics.

This paper introduces a Travel Demand Management (TDM) model in order to decrease the transportation externalities by affecting on passengers’travel choices. Thus, a bi-objective bi-modal optimization model for road pricing is developed aiming to enhance environmental and social sustainability by considering to minimize the air pollution and maximize the social welfare as its objectives. This model determines optimal prices (bus fare and car toll) and optimal bus frequency simultaneously in an integrated model. The model is based on discrete choice theory and consideres the modes’ utility functions in its formulation. The proposed model is solved by two meta-heuristic methods (Non-dominated Sorting Genetic Algorithm-II (NSGA-II), Multi-Objectives Harmony Search (MOHS)) and the numerical results of a case study in Tehran are presented. The main managerial insights resulted from this case study is that its results support the idea of “free public transportation” or subsidizing the public transport as an effective way to decrease the transport related air pollution

Increase in public concern about food safety, and the importance of factors such as food production, animal health, food safety and traceability of food products to more shows. Collecting, analyzing and interpreting information critical role in the control of food safety, animal health, public health and the environment. Data Mining is that as extracting knowledge from databases (KDD: Knowledge Discovering in Databases) is also known, is one of the most important analytical methods to identify the relationships between the various elements of the data collected in order to discover useful knowledge and decision support systems for strategic and sustainable development of the livestock, food and health industry. Mathematical modeling, quantitative analysis of data and new algorithms could identify new links between different data which, in turn, leads to competitive advantage. Tech data warehouse can be used to analyze multi-dimensional Dataset in order to make an important decision is to be used. Moreover, the prediction method can be used as a source of competitive advantage in today's business environment is considered.

In the explosive growth of Business to Business (B2B) Electronic Trades, electronic markets have received a great deal of attention recently. The obtained profit of trading in E-B2B market encourage market participants to remain in the market. Market participants consist of: sellers, buyers, and market owner. In this paper the expected profit function for each market participant has been defined in a neutral market based on double auction. Also, the model is simulated and results are shown. Linear programming in game matrix is used to exhibit the capability of the proposed model to support the decision making process. Then the model is exemplified.

Business all around the world uses different approaches to know their customers, segment them and formulate suitable strategies for them. One of these approaches is calculating the value of each customer for the company. In this paper by calculating Customer Lifetime Value (CLV) for individual customers of an online toy store named Alakdolak, three customer segments are extracted. The level of profitability for customers is identified, and finally suitable marketing strategies for each segment are developed. The results indicate that the company should increase its low price products and develop special programs for those that buy high price products and have high loyalty. Logistic regression as a data mining technique is used to present the customer defection and future purchase probability models and for each model, verification and validation is done.

Teams are defined as high performing task groups whose members are actively interdependent and share common performance objectives. This definition suggests that only those groups who perform at high levels due to their collective efforts must be considered as teams. In this paper, a model has been presented for de-termining the effects of information technology in the output of teamwork with Y structure for particular productive organizations.