Enabling Business Transformation: Applications of Artificial Intelligence in Business

Permanent URI for this collection

Browse

Recent Submissions

Now showing 1 - 5 of 5
  • Item
    A Multi-Criteria Approach toward Accelerating for Artificial Intelligence Business Ecosystems: A Perspective of AI Startups CEOs
    (2025-01-07) Han, Kyunghyun; Park, Jonghwa
    AI startups play a crucial role in introducing new ideas and technologies to the market, thereby driving the proliferation of AI. Considering the influence of AI startups within the AI business ecosystem, it is essential to support AI startups as a means of fostering economic growth. This necessitates recognizing policies related to AI startups as a critical agenda and formulating appropriate strategies to invigorate the AI business ecosystem. In other words, practical and sophisticated solutions are required to realize the potential of AI startups. This study aims to bridge the gap between rapidly advancing AI technology and the social sciences that need to support technological development. By inviting CEO and managers of major 27 AI startups in Korea, this study proposes a model for evaluating the activation of AI startups business ecosystem. Our findings indicate market demand, training AI professionals, and high-quality data is the most important factors for accelerating AI startup ecosystem. The implications of our findings underline the importance of strategic policymaking.
  • Item
    Unveiling the Dynamics of Open-Source AI Models: Development Trends, Industry Applications, and Challenges
    (2025-01-07) Sangari, Esmat; Abughoush, Khaled; Azarm, Mana
    In this study, we analyze open-source AI model development and utilization from 2012 to 2024, using data from Hugging Face and Scopus. Our findings reveal a significant surge in model development post-2020, particularly in text processing tasks, likely due to transformer model advancements. However, audio and image processing domains have grown more slowly. User engagement metrics indicate that the top 1% of models, especially in text processing, vastly outperform others, suggesting concentrated interest in specific tasks. Some popular models demonstrate versatility across tasks like image classification and reinforcement learning. The software industry leads in AI usage, followed by healthcare and education. Our study underscores the need for standardized documentation and protocols to improve model transparency and academic rigor, as many models lack clear task associations and training information. These insights provide an overview of the evolving open-model development landscape, highlighting trends, user preferences, and areas for future research and standardization.
  • Item
    An Advanced BERT-Based Commodity Classification on Amazon Online Malls Based on Consumer Cognitive Attributes
    (2025-01-07) Liu, Feng; Zheng, Qijian
    The advent of the Internet economy era has led to a surge of interest in the efficient management of e-commerce platforms. However, There is a lack of an objective method to evaluate product classification standards, helping policymakers observe their alignment with the market. To address this, we propose the BEML framework, which uses machine learning methods driven by deep learning latent space representations for standard evaluation. Our framework treats the product classification task as a context. It encodes product information using an encoder and simulates classification criteria through a machine learning classifier. Finally, it evaluates the alignment between the product market and classification standards based on the classification efficiency. Through testing 100 kinds of products on the Amazon platform in 2023, our framework evaluates the alignment between the product market and classification standards. The experimental results demonstrate that the BEML framework achieves a macro F1 score of 85.79% and a micro F1 score of 84.73%. Both exceed the current best F1 score by 83.3%, reaching a state-of-the-art level. It provides an efficient and reliable blended learning analysis paradigm for the field of technology and business studies.
  • Item
    A Human Perspective to AI-based Candidate Screening
    (2025-01-07) Vásquez-Rodríguez, Laura; Audrin, Bertrand; Michel, Samuel; Galli, Samuele; Rogenhofer, Julneth; Negro Cusa, Jacopo; Van Der Plas, Lonneke
    Skill extraction is at the core of algorithmic hiring. It is based on identifying terms commonly found in both targets (i.e., resumes and job offers), aiming at identifying a “match” or correspondence between both. This paper focuses on skill extraction from resumes, as opposed to job offers, and considers this task both from the Human Resource Management (HRM) and AI points of view. We discuss challenges identified by both fields and explain how collaboration is instrumental for a successful digital transformation of HRM. We argue that annotation efforts are an ideal example of where collaboration between both fields is needed and present an annotation effort on 46 resumes with 41 trained annotators, resulting in a total of 116 annotations. We analyze the skills extracted by multiple different systems and compare those to the skills selected by the annotators, and find that the skills extracted differ a lot in terms of length and semantic content. The skills extracted with conversational Large Language Models (LLMs) tend to be very long and detailed, other systems are very concise, whereas humans are in the middle. In terms of semantic similarity, conversational LLMs are closer to human outputs than other systems. Our analysis proposes a different perspective to understand the well-studied, but still unsolved skill extraction task. Finally, we provide recommendations for the skill extraction task that aligns with both HR and computational perspectives.