Tailored Retrieval of Health Information from the Web for Facilitating Communication and Empowerment of Elderly People

Abstract

A patient, nowadays, acquires health information from the Web mainly through a "human-to-machine" communication process with a generic search engine. This, in turn, affects, positively or negatively, his/her empowerment level and the "human-to-human" communication process that occurs between a patient and a healthcare professional such as a doctor. A generic communication process can be modelled by considering its syntactic-technical, semantic-meaning, and pragmatic-effectiveness levels and an efficacious communication occurs when all the communication levels are fully addressed. In the case of retrieval of health information from the Web, although a generic search engine is able to work at the syntactic-technical level, the semantic and pragmatic aspects are left to the user and this can be challenging, especially for elderly people. This work presents a custom search engine, FACILE, that works at the three communication levels and allows to overcome the challenges confronted during the search process. A patient can specify his/her information requirements in a simple way and FACILE will retrieve the "right" amount of Web content in a language that he/she can easily understand. This facilitates the comprehension of the found information and positively affects the empowerment process and communication with healthcare professionals. 1 INTRODUCTION According to the World Health Organization (WHO), empowerment is "a process through which people gain greater control over decisions and actions affecting their health" (WHO, 1998). It includes, as a basic step, the acquisition of health/medical information that helps patients/citizens to understand medical conditions and treatments, acquire selfconfidence to discuss them with medical professionals and, together, make the best-informed decisions (Akerkar & Bichile, 2004; Smith, 2004). The main source of health/medical information is, nowadays, the World Wide Web (or Web, for short) with the number of Web health information seekers that have been steadily increasing over the years (Pew Research Center, 2013; Taylor, 2010). Search engines are more and more used as the main tools to provide Web information. However, generic search engines do not make any distinction among the users and overload them with the amount of information. The use of a search engine for Web information retrieval entails a human-to-machine communication process between the user (e.g., a patient) and the search engine. It affects, among others, the amount, comprehension and use of the found information. This, in turn, may facilitate or complicate the humanto-human communication process between patients and healthcare professionals such as doctors (Smith, 2004). Communication processes have been modelled in various ways in the past decades. One of the most famous communication model is the one introduced by Morris in relation to its theory of signs (Morris, 1938). It is made up of three levels, i.e., syntactic, semantic and pragmatic. It is paired up by another communication model introduced by Shannon...

On-Line Retrieval of Health Information Based on Language Complexity, Information Customization and Information Quality

“…The above considerations about language complexity, information customization and information quality have led us to create a mapping between the user requirements and the schema.org elements for the non-expert and expert user categories. This mapping expands the one presented in [4,24] and is reported in Table 1.…”

“…Some of the information presented in this paper is based on a previous work [24]. The present paper, however, extends the previous study by adopting a newer dataset, from 2019, of structured data and extracting a higher number of health-lifesci.schema.org and schema.org quadruples so to provide the user with a richer set of information.…”

“…The mapping between user requirements and schema.org elements, shown in the previous Section, has been used to build FACILE in order to provide experts and non-experts with the proper Web contents in terms of language complexity, information customization and information quality. It completes the previous versions of the system, the first one considering only the language level [31] and the second one taking into account only the information customization and quality [24]. Moreover, the algorithm used to provide the custom information has been completely revised with respect to the previous versions.…”