Innovation Management is a key process for mapping a technological domain and managing technological capabilities, either for companies, research institutions and policy makers. It allows managers and decision makers to anticipate trends for accurate forecast and effective foresight. A technological innovation starts with the ideation phase, in which researchers, inventors, and engineers (innovation actors) design and develop a novel technology. The main discipline that studies and attempts to standardize the technology design and development processes is Engineering Design (ED). The innovation actors operate in these phases using ED practices and concepts, such as technology functions, users, technical problems and so on. Moreover, they are pushed by industrial and academic system, to produce patents, scientific publications, and company technical documents to describe their inventive steps. These documents are an unavailable source of knowledge that reflects the cognitive process of innovation actors and encapsulate the main technical concepts of ED. Among these sources, patents are the widest technical open access database used in literature and in practice for studying technological phenomena. Nowadays, Text mining and Natural Language Processing (NLP) provides new methods for the analysis of patent texts. NLP is a branch of text mining for the automatic processing of the human language (natural language in jargon) in written form. The application of NLP for the analysis of technological information is called Tech Mining. However, most Tech Mining methods used in literature do not consider the meaning of the textual information and the ED expert knowledge generated during the process of research, design and develop. My Ph.D research is focused on demonstrating that the identification and mapping of the ED knowledge from the text of patents may enhance the innovation management process. First, I review the literature to provide a clear picture of state-of-the-art in Tech Mining focused on ED concepts. Then, I present a Tech Mining system for identifying ED concepts (i.e., technological terms, technical problems, solution to the problems and advantageous effects) from patents. Finally, I prose an approach to study the trends of technological evolution using the ED concepts. My study delineates valid Tech Mining tools that can be integrated in any text analysis pipeline to support academics and companies in investigating a technological domain. This tool allows organizations to focus on value-added activities of technological forecasting and management process.

Engineering Tech Mining: mixing Engineering Design and Tech Mining for Innovation Management

GIORDANO, VITO
2022

Abstract

Innovation Management is a key process for mapping a technological domain and managing technological capabilities, either for companies, research institutions and policy makers. It allows managers and decision makers to anticipate trends for accurate forecast and effective foresight. A technological innovation starts with the ideation phase, in which researchers, inventors, and engineers (innovation actors) design and develop a novel technology. The main discipline that studies and attempts to standardize the technology design and development processes is Engineering Design (ED). The innovation actors operate in these phases using ED practices and concepts, such as technology functions, users, technical problems and so on. Moreover, they are pushed by industrial and academic system, to produce patents, scientific publications, and company technical documents to describe their inventive steps. These documents are an unavailable source of knowledge that reflects the cognitive process of innovation actors and encapsulate the main technical concepts of ED. Among these sources, patents are the widest technical open access database used in literature and in practice for studying technological phenomena. Nowadays, Text mining and Natural Language Processing (NLP) provides new methods for the analysis of patent texts. NLP is a branch of text mining for the automatic processing of the human language (natural language in jargon) in written form. The application of NLP for the analysis of technological information is called Tech Mining. However, most Tech Mining methods used in literature do not consider the meaning of the textual information and the ED expert knowledge generated during the process of research, design and develop. My Ph.D research is focused on demonstrating that the identification and mapping of the ED knowledge from the text of patents may enhance the innovation management process. First, I review the literature to provide a clear picture of state-of-the-art in Tech Mining focused on ED concepts. Then, I present a Tech Mining system for identifying ED concepts (i.e., technological terms, technical problems, solution to the problems and advantageous effects) from patents. Finally, I prose an approach to study the trends of technological evolution using the ED concepts. My study delineates valid Tech Mining tools that can be integrated in any text analysis pipeline to support academics and companies in investigating a technological domain. This tool allows organizations to focus on value-added activities of technological forecasting and management process.
13-dic-2022
Italiano
information retrieval
natural language processing
patent analysis
tech mining
technology analysis
text mining
Fantoni, Gualtiero
Chiarello, Filippo
File in questo prodotto:
File Dimensione Formato  
Doctoral_Thesis___Vito_Giordano.pdf

embargo fino al 15/12/2025

Dimensione 4.49 MB
Formato Adobe PDF
4.49 MB Adobe PDF
Report_Attivita___Vito_Giordano.pdf

non disponibili

Dimensione 406.95 kB
Formato Adobe PDF
406.95 kB Adobe PDF
Sintesi_Tesi___Vito_Giordano.pdf

embargo fino al 15/12/2025

Dimensione 83.23 kB
Formato Adobe PDF
83.23 kB Adobe PDF

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14242/215502
Il codice NBN di questa tesi è URN:NBN:IT:UNIPI-215502