1,080 research outputs found
Blending brainstorming with industrial experts and automatic data-driven analysis
No full textThe goal of ASSETS+ project is developing a strategy to upskill and reskill students and professionals in Defence sector, starting from identifying and mapping the most important technologies and exploring their impact on skills and human capabilities. The technological focus of the project is on Robotics, Autonomous Systems, Artificial Intelligence (AI), Cybersecurity, and C4ISTAR
Foreseeing Future Applications of Generative Large Language Models: A Data-Driven Case Study on the Emergence of ChatGPT
No full textWe study the evolving role of generative Large Language Models (LLMs). We develop a data-driven approach to collect and analyse tasks that users are asking to generative LLMs. Thanks to the focus on tasks this paper contributes to give a quantitative and granular understanding of the potential influence of LLMs in different business areas. Utilizing a dataset comprising over 3.8 million tweets, we identify and cluster 31,747 unique tasks, with a specific case study on ChatGPT. To reach this goal, the proposed method combines two Natural Language Processing (NLP) Techniques, Named Entity Recognition (NER) and BERTopic algorithms. The combination makes it possible to collect granular tasks of LLMs (NER) and clusters them in business areas (BERTopic). Our findings reveal a wide spectrum of applications, from programming assistance to creative content generation, highlighting LLM’s versatility. The analysis highlighted six emerging areas of application for ChatGPT: human resources, programming, social media, office automation, search engines, education. The study also examines the implications of these findings for innovation management, proposing a research agenda to explore the intersection of the identified areas, with four stages of the innovation process: idea generation, screening/idea selection, development, and diffusion/sales/marketing
How can Generative LLMs support Designers? Escaping the “Brainstorming Assistant” Fixation
No full textIn this research paper we explore the role of Generative Large Language Models (Gen-LLMs) in supporting the product design process, moving beyond their conventional application as mere brainstorming assistants. Through an automated review of 15,355 design-related scientific articles, we identify and classify language-based tasks Gen-LLMs can enhance across various stages of product design. Employing a novel methodological approach that combines Named Entity Recognition with qualitative task classification, we classify these tasks into three primary functions: generation, evaluation, and description, subsequently mapping them to the corresponding phases of the design process: Problem Definition, Conceptual Design, Embodiment Design, and Detailed Design. Our findings reveal that while the generative capabilities of Gen-LLMs have been the primary focus, their potential to augment evaluative and descriptive tasks is significant and underexplored. By highlighting the diverse applications of Gen-LLMs, this study challenges the existing research and practice paradigm, advocating for a broader exploration of Gen-LLMs' capabilities. It underscores the necessity of balancing automation with human-centric design values and proposes a research agenda for further integrating Gen-LLMs into design practices, thus enriching both the theoretical and practical aspects
Unveiling the Uses of Generative Large Language: a Case Study on Twitter Data About ChatGPT
No full textThis study examines the evolving role of generative Large Language Models (LLMs). Using a data-driven approach, we collect and analyze the tasks users direct to these models, offering a quantitative and detailed perspective on their potential impact across various business domains. Drawing on a dataset of over 3.8 million tweets, we identify and categorize 31,747 unique tasks, including a focused case study on ChatGPT. Our methodology integrates two Natural Language Processing (NLP) techniques—Named Entity Recognition (NER) and BERTopic—to capture granular LLM-related tasks (NER) and organize them into business clusters (BERTopic). The findings reveal a broad range of applications, from programming assistance to creative content generation, underscoring the versatility of LLMs. The analysis highlights six emerging application areas for ChatGPT: human resources, programming, social media, office automation, search engines, and education. The study further considers the implications for innovation management, proposing a research agenda that explores how these identified areas align with the four stages of the innovation process: idea generation, screening/selection, development, and diffusion/sales/marketing
Technical Sentiment Analysis: Measuring Advantages and Drawbacks of New Products Using Social Media
No full textIn recent years, social media have become ubiquitous and important for social networking and content sharing. Moreover, the content generated by these websites remains largely untapped. Some researchers proved that social media have been a valuable source to predict the future outcomes of some events such as box-office movie revenues or political elections. Social media are also used by companies to measure the sentiment of customers about their brand and products.
This work proposes a new social media based model to measure how users perceive new products from a technical point of view. This model relies on the analysis of advantages and drawbacks of products, which are both important aspects evaluated by consumers during the buying decision process. This model is based on a lexicon developed in a related work (Chiarello et. al, 2017) to analyse patents and detect advantages and drawbacks connected to a certain technology.
The results show that when a product has a certain technological complexity and fuels a more technical debate, advantages and drawbacks analysis is more efficient than sentiment analysis in producing technical-functional judgements
SSH researchers make an impact differently. Looking at public research from the perspective of users
No full textWith the rise of the impact assessment revolution, governments and public opinion have started to ask researchers to give evidence of their impact outside the traditional audiences, i.e. students and researchers. There is a mismatch between the request to demonstrate the impact and the current methodologies for impact assessment. This mismatch is particularly worrisome for the research in Social Sciences and Humanities. This paper gives a contribution by examining systematically a key element of impact, i.e. the social groups that are directly or indirectly affected by the results of research. We use a Text mining approach applied to the Research Excellence Framework (REF) collection of 6,637 impact case studies in order to identify social groups mentioned by researchers. Differently from previous studies, we employ a lexicon of user groups that includes 76,857 entries, which saturates the semantic field, permits the identification of all users and opens the way to normalization. We then develop three new metrics measuring Frequency, Diversity and Specificity of user expressions. We find that Social Sciences and Humanities exhibit a distinctive structure with respect to frequency and specificity of users
AI-Based Knowledge Extraction from the Bioprinting Literature for Identifying Technology Trends
No full textB
ioprinting is a rapidly evolving field, as represented by the exponential growth of articles and reviews
published each year on the topic. As the number of publications increases, there is a need for an automatic tool
that can help researchers do more comprehensive literature analysis, standardize the nomenclature, and so
accelerate the development of novel manufacturing techniques and materials for the field. In this context, we
propose an automatic keyword annotation model, based on Natural Language Processing (NLP) techniques, that
can be used to find insights in the bioprinting scientific literature. The approach is based on two main data
sources, the abstracts and related author keywords, which are used to train a composite model based on (i) an
embeddings part (using the FastText algorithm), which generates word vectors for an input keyword, and (ii) a
classifier part (using the Support Vector Machine algorithm), to label the keyword based on its word vector into
a manufacturing technique, employed material, or application of the bioprinted product. The composite model
was trained and optimized based on a two-stage optimization procedure to yield the best classification per-
formance. The annotated author keywords were then reprojected on the abstract collection to both generate a
lexicon of the bioprinting field and extract relevant information, like technology trends and the relationship
between manufacturing-material-application. The proposed approach can serve as a basis for more complex
NLP-related analysis toward the automated analysis of the bioprinting literature
A simple and fast method for Named Entity context extraction from patents
No full textThe process of extracting relevant technical information from patents or technical literature is as valuable as it is challenging. It deals with highly relevant information extraction from a corpus of documents with particular structure, and a mix of technical and legal jargon. Patents are the wider free source of technical information where homogeneous entities can be found. From a technical perspective the approaches refer to Named Entity Recognition (NER) and make use of Machine Learning techniques for Natural Language Processing (NLP). However, due to the large amount of data, to the complexity of the lexicon, the peculiarity of the structure and the scarcity of the examples to be used to feed the machine learning system, new approaches should be studied. NER methods are increasing their performances in many contexts, but a gap still exists when dealing with technical documentation. The aim of this work is to create an automatic training sets for NER systems by exploiting the nature and structure of patents, an open and massive source of technical documentation. In particular, we focus on collecting the context where users of the invention appear within patents. We then measure to which extent we achieve our goal and discuss how much our method is generalizable to other entities and documents
How Large Language Models Can Change Innovation Management?
No full textThis study explores how generative Large Language Models (LLMs) can transform innovation management by shaping and enhancing practices across various business domains. Employing a data-driven approach, we gather and analyze the tasks users direct to these models, providing a quantitative, detailed perspective on their potential influence. Drawing from a dataset of over 3.8 million tweets, we identify and categorize 31,747 unique tasks, including a focused case study on ChatGPT. To achieve this, we combine two Natural Language Processing (NLP) techniques—Named Entity Recognition (NER) and BERTopic—thereby capturing the granular tasks associated with LLMs (NER) and grouping them into coherent business clusters (BERTopic).
Our findings uncover a broad spectrum of applications, ranging from programming assistance to creative content generation, highlighting the versatility of LLMs. In particular, the analysis points to six emerging areas for ChatGPT: human resources, programming, social media, office automation, search engines, and education. We then connect these areas to the four stages of the innovation process—idea generation, screening/selection, development, and diffusion/sales/marketing—proposing a research agenda that integrates LLM-driven insights with key innovation management activities
Stakeholder Engagement in the Design of Artificial Intelligence
No full textThis article investigates stakeholder engagement in research studies on the design and development of Artificial Intelligence (AI). The research investigates the innovation and design literature, exploring the implications of AI within the rapidly evolving technology and business landscape. In this rapid evolution of AI, various cases emerged in practice in which AI applications caused negative impacts, such as harm to people. Acknowledging the intrinsic focus of AI research studies on optimizing computational objectives, this research highlights the often overlooked diverse needs of specific stakeholders, which can lead to significant challenges, notably racial and gender biases. These cases underscore the need for a thorough examination of stakeholder engagement in the current research studies on the design and development of AI. Therefore, this study includes a data-driven, in-depth analysis of stakeholder engagement in the digital design and development of AI. The article discusses several practices to engage stakeholders in different design activities in AI development. This article aims to foster a more stakeholder-centric, inclusive, and responsible AI development landscape. The research enhances the ongoing discourse on AI’s role in design and innovation management practices
- …
