Software Technology to Develop Large-Scale Self-Adaptive Systems: Accelerating Agent-Based Models and Fuzzy Cognitive Maps via CUDA

2023 | HICSS | Citations: 0

Authors: Ghumrawi, Kareem; Ha, Kim; Beerman, Jack; Rudie, John-David; Giabbanelli, Philippe

Abstract: Agent-Based Models (ABMs) have long served to study self-adaptive systems and th ... Expand

Abstract: Agent-Based Models (ABMs) have long served to study self-adaptive systems and the emergence of population-wide patterns from simple rules applied to individuals. Recently, the rules for each agent have been expressed using a Fuzzy Cognitive Map (FCM), which is elicited from a subject-matter expert. This provides a transparent and participatory process to externalize the `mental model' of an expert and directly embed it into agents. However, software technology has been lacking to support such hybrid ABM/FCM models at scale, which has drastically limited the scope of applications and the ability of researchers to study emergent phenomena over large populations. In this paper, we designed and implemented the first open-source library that automatically accelerates ABM/FCM models by leveraging the CUDA cores available in a Graphical Processing Unit. We demonstrate the correctness and scaling of our library on a case study as well as across different networks representing agent interactions. Collapse

Topics: social network computer hardware Python vendor lock-in parallel computing

Methods: computational algorithm simulation case study cross sectional research

Heterogeneous Energy-aware Load Balancing for Industry 4.0 and IoT Environments

2022 | ACM Transactions on Management Information Systems | Citations: 0

Authors: Ahmed, Usman; Lin, Jerry Chun-Wei; Srivastava, Gautam

Abstract: With the improvement of global infrastructure, Cyber-Physical Systems (CPS) have ... Expand

Abstract: With the improvement of global infrastructure, Cyber-Physical Systems (CPS) have become an important component of Industry 4.0. Both the application as well as the machine work together to improve the task of interdependencies. Machine learning methods in CPS require the monitoring of computational algorithms, including adopting optimizations, fine-tuning cyber systems, improving resource utilization, as well as reducing vulnerability and also computation time. By leveraging the tremendous parallelism provided by General-Purpose Graphics Processing Units (GPGPU) as well as OpenCL, it is possible to dramatically reduce the execution time of data-parallel programs. However, when running an application with tiny amounts of data on a GPU, GPU resources are wasted because the program may not be able to fully utilize the GPU cores. This is because there is no mechanism for kernels to share a GPU due to the lack of OS support for GPUs. Optimal device selection is required to reduce the high power of the GPU. In this paper, we propose an energy reduction method for heterogeneous clustering. This study focuses on load balancing; resource-aware processor selection based on machine learning is performed using code features. The proposed method identifies energy-efficient kernel candidates (from the employment pool). Then, it selects a pair of kernel candidates from all possibilities that lead to a reduction in both energy consumption as well as execution time. Experimental results show that the proposed kernel approach reduces execution time by 2.23 times compared to a baseline scheduling system. Experiments have also shown that the execution time is 1.2 times faster than state-of-the-art approaches. Collapse

Topics: production management computer hardware database system internet of things mobile system

Methods: computational algorithm machine learning experiment random forest classification gradient boosting

Strategic business value from big data analytics: An empirical analysis of the mediating effects of value creation mechanisms

2022 | Information & Management | Citations: 3

Authors: Elia, Gianluca; Raguseo, Elisabetta; Solazzo, Gianluca; Pigni, Federico

Abstract: Big data are a prominent source of value capable of generating competitive advan ... Expand

Abstract: Big data are a prominent source of value capable of generating competitive advantage and superior business performance. This paper represents the first empirical investigation of the theoretical model proposed by Grover et al. (2018), considering the mediating effects of four value creation mechanisms on the relationship between big data analytics capabilities (BDAC) and four value targets. The four value creation mechanisms investigated (the source of the value being pursued) are transparency, access, discovery, and proactive adaptation, while the four value targets (the impacts of the value creation process) are organization performance, business process improvement, customer experience and market enhancement, and product and service innovation. The proposed empirical validation of Grover et al.’s (2018) model adopts an econometric analysis applied to data gathered through a survey involving 256 BDA experts. The results reveal that transparency mediates the relationship for all the value targets, while access and proactive adaptation mediate only in case of some value targets, and discovery does not have any mediating effect. Theoretical and practical implications are discussed at the end of the paper. Collapse

Topics: value creation big data innovation management organizational productivity business process improvement

Methods: survey theory development company material exploratory data analysis longitudinal research

Theories: theory of economic development dynamic capabilities theory resource based view of the firm

Human-Machine Hybrid Decision Making with Applications in Auditing

2022 | HICSS | Citations: 0

Authors: Hooshangi, Sara; Sibdari, Soheil

Abstract: The decision making process in a variety of organizations faces substantial chan ... Expand

Abstract: The decision making process in a variety of organizations faces substantial changes, largely as a result of advances in information technology and artificial intelligence (AI). A considerable number of decisions that were traditionally made by humans, are now made by machines. Consequently, many jobs that were held by experts in some fields are now occupied by data scientists who can build AI algorithms. In this paper, we address this change in work environments and suggest an innovative process for hybrid decision making between humans and machines. We focus on the auditing profession, but our method can be used in other human-intensive and critical fields such as healthcare, financial services public sectors, and humanitarian organizations. Collapse

Topics: decision making artificial intelligence usability human computer interaction decision support system

Methods: survey qualitative interview experiment personal interview literature study

Multicategory choice modeling with sparse and high dimensional data: A Bayesian deep learning approach

2022 | Decision Support Systems | Citations: 1

Authors: Xia, Feihong; Chatterjee, Rabikar

Abstract: The availability of sparse and high dimensional consumer shopping data poses a c ... Expand

Abstract: The availability of sparse and high dimensional consumer shopping data poses a challenge for researchers for accurate and efficient analysis. While deep learning models can handle such data, most of the results from these models are uninterpretable. This greatly limits their value in applications aimed at better understanding mul­ ticategory shopping behavior and assisting managerial decision-making. Thus, a new approach is needed to analyze high dimensional consumer shopping data in an efficient and interpretable manner, with the potential to greatly strengthen decision support systems for managers especially at large consumer packaged goods firms and giant retailers. We propose a Bayesian deep learning approach based on Variational Autoencoder (VAE) designed to efficiently capture often complex and multifaceted interrelationships across items in the shopping baskets, in the form of substitute and complementary cross effects (which are observable and controllable) and coincidences (which are not). The benefits of the proposed approach are empirically demonstrated by applying our model to a high dimensional supermarket shopping dataset covering very large numbers of products at the stock-keeping unit (SKU) level, households, and shopping trips. We discuss implications for managerial decision-making and identify promising research directions. Collapse

Topics: marketing management decision making open source accounting decision support system

Methods: artificial neural network machine learning deep learning data transformation factor analysis

Music intelligence: Granular data and prediction of top ten hit songs

2021 | Decision Support Systems | Citations: 1

Authors: Kim, Seon Tae; Oh, Joo Hee

Abstract: In the music market, superstars significantly dominate the market share, while p ... Expand

Abstract: In the music market, superstars significantly dominate the market share, while predicting the top hit songs is notoriously difficult. The music intelligence technology, retrieving and utilizing granular acoustic features of songs, provides opportunities to improve the prediction of top hit songs. Using data on 6209 unique songs that appeared in the weekly Billboard Hot 100 charts from 1998 to 2016, especially acoustic features provided by Spotify, we investigate empirically how the top-10-hit-songs likelihood prediction is improved by acoustic fea­ tures. We find that some acoustic features (e.g., danceability, happiness, and some metrics of timbre and pitch) significantly improve the model’s ability to predict the top-10-hit-songs probability. These results suggest that the granular data, provided by the music intelligence technology, carries a substantial predictive value in the era of online music streaming. Collapse

Topics: digital service Spotify music streaming service logistics management internet technology

Methods: computational algorithm machine learning descriptive statistic generalized linear model McNemar's test

Artificial intelligence capability: Conceptualization, measurement calibration, and empirical study on its impact on organizational creativity and firm performance

2021 | Information & Management | Citations: 0

Authors: Mikalef, Patrick; Gupta, Manjul

Abstract: Artificial intelligence (AI) has been heralded by many as the next source of bus ... Expand

Abstract: Artificial intelligence (AI) has been heralded by many as the next source of business value. Grounded on the resource-based theory of the firm and on recent work on AI at the organizational context, this study (1) identifies the AI-specific resources that jointly create an AI capability and provides a definition, (2) develops an instrument to capture the AI capability of the firms, and (3) examines the relationship between an AI capability and organizational creativity and performance. Findings empirically support the suggested theoretical framework and corresponding instrument and provide evidence that an AI capability results in increased organizational creativity and performance. Collapse

Topics: artificial intelligence organizational productivity information technology capability organizational value organizational context

Methods: survey design psychometrics data transformation theory development partial least squares path modeling

Theories: transaction cost economics resource based view of the firm

Upgrading Products based on Existing Dominant Competitors

2021 | HICSS | Citations: 0

Authors: Wijayanto, Heri; Thamrin, Syauki; Chen, Arbee L.P.

Abstract: In the Industry 4.0 era, manufacturers compete to produce better products that a ... Expand

Abstract: In the Industry 4.0 era, manufacturers compete to produce better products that are expected to satisfy a larger number of customers. We propose a recommendation system for upgrading products considering user preferences. This approach is based on the dominating regions of dominant competitors. The dominating region represents the estimation of the number of potential customers. However, examining overlapped dominating regions for a high dimensional space is NP-hard. We propose a novel method named TDRDFS which constructs a Dominant Graph of Intersection skyline points (DGI) for modeling the dominating regions. Our experiments show that TDRDFS significantly reduces computation. Based on our approach, product vendors are able to determine the strategy of upgrading products easily. Collapse

Topics: recommender system MapReduce parallel computing personal computer smartphone

Methods: computational algorithm experiment literature study

Efficient Model Points Selection in Insurance by Parallel Global Optimization Using Multi CPU and Multi GPU

2020 | Business & Information Systems Engineering | Citations: 0

Authors: Ferreiro-Ferreiro, Ana Maria; García-Rodríguez, José Antonio; Souto, Luis A.; Vázquez, Carlos

Abstract: In the insurance sector, Asset Liability Management refers to the joint manageme ... Expand

Abstract: In the insurance sector, Asset Liability Management refers to the joint management of the assets and liabilities of a company. The liabilities mainly consist of the policies portfolios of the insurance company, which usually contain a large amount of policies. In the article, the authors mainly develop a highly efficient automatic generation of model points portfolios to represent much larger real policies portfolios. The obtained model points portfolio must retain the market risk properties of the initial portfolio. For this purpose, the authors propose a risk measure that incorporates the uncertain evolution of interest rates to the portfolios of life insurance policies, following Ferri (Optimal model points portfolio in life, 2019, arXiv:1808.00866). This problem can be formulated as a minimization problem that has to be solved using global numerical optimization algorithms. The cost functional measures an appropriate distance between the original and the model point portfolios. In order to solve this problem in a reasonable computing time, sequential implementations become prohibitive, so that the authors speed up the computations by developing a high performance computing framework that uses hybrid architectures, which consist of multi CPUs together with accelerators (multi GPUs). Thus, in graphic processor units (GPUs) the evaluation of the cost function is parallelized, which requires a Monte Carlo method. For the optimization problem, the authors compare a metaheuristic stochastic differential evolution algorithm with a multi path variant of hybrid global optimization Basin Hopping algorithms, which combines Simulated Annealing with gradient local searchers (Ferreiro et al. in Appl Math Comput 356:282–298, 2019a). Both global optimizers are parallelized in a multi CPU together with a multi GPU setting. Collapse

Topics: biometric parallel computing computer hardware accounting application programming interface

Methods: computational algorithm Monte Carlo simulation optimization model design artifact testing experiment

Extracting Maritime Traffic Networks from AIS Data Using Evolutionary Algorithm

2020 | Business & Information Systems Engineering | Citations: 0

Authors: Filipiak, Dominik; Węcel, Krzysztof; Stróżyna, Milena; Michalak, Michał; Abramowicz, Witold

Abstract: The presented method reconstructs a network (a graph) from AIS data, which refle ... Expand

Abstract: The presented method reconstructs a network (a graph) from AIS data, which reflects vessel traffic and can be used for route planning. The approach consists of three main steps: maneuvering points detection, waypoints discovery, and edge construction. The maneuvering points detection uses the CUSUM method and reduces the amount of data for further processing. The genetic algorithm with spatial partitioning is used for waypoints discovery. Finally, edges connecting these waypoints form the final maritime traffic network. The approach aims at advancing the practice of maritime voyage planning, which is typically done manually by a ship’s navigation officer. The authors demonstrate the results of the implementation using Apache Spark, a popular distributed and parallel computing framework. The method is evaluated by comparing the results with an on-line voyage planning application. The evaluation shows that the approach has the capacity to generate a graph which resembles the real-world maritime traffic network. Collapse

Topics: Apache Spark parallel computing computer hardware

Methods: computational algorithm cumulative sum control chart experiment machine learning optimization model

Theories: graph theory