Normal view MARC view ISBD view

Chapter A Natural Language Processing approach to measure expertise in the Delphi-based scenarios

By:

Calleo, Yuri [auth]

Contributor(s):

Material type: Text

TextLanguage: English Series: Publication details: Florence Firenze University Press, Genova University Press 2023Description: 1 electronic resource (6 p.)Content type:

text

Media type:

computer

Carrier type:

online resource

ISBN:

979-12-215-0106-3.29
9791221501063

Subject(s):

Online resources:

Summary: The Delphi-based scenarios (DBS) development implies the assumption of different choices, through the gathering of information and the assessment of alternative resolutions (Panpatte and Takale, 2019). During the last decades, the spread of environmental hazards has increased quickly, much to request different responses in order to develop a sustainable future for humanity planning the present (McMichael and Lindgren, 2011). Since the DBS is a creative process (Nowack et al., 2011), different figures are selected to make choices, including academics, stakeholders and citizens. However, one of the main challenges remains the measurement of expertise, in fact, during the process, the experts should be assessed based on their competences in order to avoid any conflict in the final results and, eventually, weigh their answers. In recent years, some contributions adopted the self-assessments for the experts' evaluation (Sossa et al., 2019), but many issues still remain (such as strong subjectivity and cognitive biases which produce over or underestimation). We develop a new method to estimate the expertise by using Natural Language Processing to acquire information, extracting the contributions of experts in each topic. First, starting from a draft list of selected experts, we identify the category of reference (e.g., academia, industry, local authority, citizens etc.). We build a data repository with the personal pages (URLs) of each expert to then use Python to extract from the URLs, the number of contributions related to a keyword, different for each category (e.g., publications for academics, reports and projects for stakeholders etc.). Finally, we proceed adopting a coefficient of production with a weighted sum of the results. To practically demonstrate our approach, we applied this method to a cohort of known experts, part of the "Smart control of the climate resilience" (SCORE) H2020 European project to estimate their expertise in specific areas.

Item type:

Tags from this library: No tags from this library for this title. Log in to add tags.

Average rating: 0.0 (0 votes)

Holdings
Item type	Home library	Collection	Call number	Materials specified	Status	Date due	Barcode
Electronic-Books	OPJGU Sonepat- Campus	E-Books Open Access			Available

Open Access Unrestricted online access star

The Delphi-based scenarios (DBS) development implies the assumption of different choices, through the gathering of information and the assessment of alternative resolutions (Panpatte and Takale, 2019). During the last decades, the spread of environmental hazards has increased quickly, much to request different responses in order to develop a sustainable future for humanity planning the present (McMichael and Lindgren, 2011). Since the DBS is a creative process (Nowack et al., 2011), different figures are selected to make choices, including academics, stakeholders and citizens. However, one of the main challenges remains the measurement of expertise, in fact, during the process, the experts should be assessed based on their competences in order to avoid any conflict in the final results and, eventually, weigh their answers. In recent years, some contributions adopted the self-assessments for the experts' evaluation (Sossa et al., 2019), but many issues still remain (such as strong subjectivity and cognitive biases which produce over or underestimation). We develop a new method to estimate the expertise by using Natural Language Processing to acquire information, extracting the contributions of experts in each topic. First, starting from a draft list of selected experts, we identify the category of reference (e.g., academia, industry, local authority, citizens etc.). We build a data repository with the personal pages (URLs) of each expert to then use Python to extract from the URLs, the number of contributions related to a keyword, different for each category (e.g., publications for academics, reports and projects for stakeholders etc.). Finally, we proceed adopting a coefficient of production with a weighted sum of the results. To practically demonstrate our approach, we applied this method to a cohort of known experts, part of the "Smart control of the climate resilience" (SCORE) H2020 European project to estimate their expertise in specific areas.