Guest editorial: Artificial intelligence and machine learning in business and management

Artificial intelligence and machine learning in business and managementArtificial intelligence (AI) is the enhancement of computers so that they can perform complex activities and tasks in ways that can be labeled as intelligent.Machine learning (ML) is a branch of AI where algorithms are used to learn from data to make decisions or predictions.Timely research subjects in AI and ML are: reinforcement learning, ethics in AI, quantum computing, convergence of AI and other emerging technologies, facial recognition, biased data, neural networks, socioeconomic models, deep learning and privacy protection (Alpaydin, 2014;Jiang, 2021;Shalev-Shwartz and Ben-David, 2014).AI and ML are drawing considerable attention these days.An already large and fastgrowing literature covers a wide range of real or soon-to-materialize applications, from dealing with autism and detecting breast cancer to selecting job candidates, setting legal sanctions and deciding upon conditional liberations.Managers, among many others, are now in need of a compass that would allow them to navigate lucidly through the hype (Bryson et al., 2021;La Torre et al., 2021; Jammeli et al., 2021 and the references therein).This special issue aims to start developing such a compass.It includes novel contributions and innovative ways of applying AI and ML techniques to business.It also covers most generic managerial tasks, assessing for each one the extent of the support AI and ML can bring and the consequences on managerial practices.In Al Janabi (2021), the author examines, from a commodity portfolio manager's perspective, the performance of liquidity adjusted risk modeling in assessing the market risk parameters of a large commodity portfolio and in obtaining efficient and coherent portfolios under different market circumstances.Using reinforcement machine learning techniques, the implemented market risk modeling algorithm and investment portfolio analytics can simultaneously handle risk-return characteristics of commodity investments under regular and crisis market settings, besides considering the particular effects of the time-varying liquidity constraints on the multiple-asset commodity portfolios.In Fadaei PellehShahi et al. (2021), the authors propose a method which combines deep learning methods (the recurrent neural network and Markov chain ones, in particular) to predict the final status of an ongoing process or a subsequent activity in a process.While semistructured business processes cannot be predicted by formal analytical methods, AI can be successful at it.The proposed method applies the BestFirst algorithm for the search section and the CFSSUBSETEVAL algorithm for the feature comparison section.The study focuses on the prediction systems of social insurance and presents a method which is cost efficient in providing real-world results based on past history of an event.In Khashei and Chahkotahi (2021), the authors propose a linear optimal weighting estimator (LOWE) algorithm to find the desired weight of components in a global noniterative universal manner.Although it can be generally demonstrated that the performance of the proposed weighting technique will not be worse than the metaheuristic algorithm, its performance is also practically evaluated for real-world data sets.Empirical results indicate that the accuracy of the LOWE-based parallel hybrid model is significantly better than metaheuristic as well as simple average (SA) based