A narrative-based reasoning with applications in decision support for social service organizations

Abstract

Nowadays, there is an increasing demand for incorporating unstructured narratives in decision support for knowledge-intensive industries such as healthcare and social service organizations. However, most of the current research on decision support systems (DSS) mainly focused on dealing with structured data and are inadequate to dealing with unstructured narratives such as clients’ records and stories. This paper presents a narrative-based reasoning (NBR) algorithm which incorporates the technologies of knowledge-based system (KBS), computational linguistics, and artificial intelligence (AI) for automatic processing unstructured narratives and inferring useful knowledge for decision support. A NBR enabled DSS has been built and was evaluated through a series of experiments conducted in early intervention of mental health of a social service company in Hong Kong. The performance of NBR was measured based on recall and precision and encouraging results were obtained. High recall and precision are achieved in the reasoning of unstructured data, and high recall is achieved for the association analysis. The results show that it is possible for inferring recommendations for problem solving from unstructured narratives automatically. Based on the approach, it helps to support knowledge workers with reliable suggestions on decision making so as to increase the quality of their solutions.

Introduction

Mental health problems impact seriously to the society. There is an increasing psychological morbidity of people which is illustrated by an increase in the incidence of suicide and the risk of developing depression, increasing rate of drug abuse and substance abuse, and drug related offending (Wang, Cheung, Lee, & Kwok, 2007b). However, mental health care service providers are facing growing challenge owing to the diverse environmental changes such as increasing competition, constraining budget, high complication of social problems (Ferns, 1995, Savage, 1987). The increasing trend of psychological morbidity certainly adds burden to the mental health care providers for offering timely and quality services so as to maintain the health of the community.

Moreover, changes in social welfare policy, in particular the subvention system and quality assurance mechanisms have posted tremendous demands on the mental health care providers, both in terms of service quantities and qualities. There is lacking psychiatric psychosis professional such as doctors, nurse and social workers. In some public hospitals, the social worker and client ratio is 1–100 and the nurse and client ratio is 1–50. Traditional approaches in mental health care, which social workers’ learning of practice wisdom depends on face-to-face sharing, may not be always available now. Due to the board range of knowledge and experience in mental health care, some researchers (e.g. Carlson, 1999, Kolbo and Wahington, 1999, Schoech, 1999) mentioned that the application of decision support system (DSS) mimics the processes of performing task at skill levels comparable to human experts.

DSS is an infrastructure and enabling technology for enhancing decision making process (Antony & Santhanam, 2007). The core components of DSS are knowledge base and reasoning mechanisms (Huang, 2009). The experience of knowledge workers is assimilated and stored in the knowledge base or knowledge repository in a certain format or schema. Moreover, the repository stores the knowledge invoked in prior decisions and retains the rules, policies and standard procedures of an organization (Hine & Goul, 1998). The inference or reasoning mechanisms makes use the knowledge stored to deduce suggestions to a given problem (Chau & Albermani, 2002). Based on DSS, the valuable knowledge that resides within individuals is thus identified and disseminated throughout the organization (Tan & Platts, 2004).

A number of DSSs have been built and applied to various areas which include performance assessment (Ammar et al., 2004, Wang, 2005, Wang et al., 2007a), commercial loan underwriting (Kumra, Stein, & Assersohn, 2006), logistics strategy design (Chow, Choy, Lee, & Chan, 2005), farm productivity (Pomar & Pomar, 2005), mergers and acquisitions (Wen et al., 2005a, Wen et al., 2005b), defense budget planning (Wen et al., 2005a, Wen et al., 2005b), earthquake design (Berrais, 2005), system dynamics (Yim, Kim, Kim, & Kwahk, 2004), conveyor equipment selection (Fonseca, Uppal, & Greene, 2004), customer service management (Cheung, Lee, Wang, Chu, & To, 2003), etc. However, most of them are dealing with structured information. The use of unstructured narratives for decision support has received relatively little attention.

In knowledge-intensive organizations such as healthcare and social service organizations, most of the knowledge resided in the unstructured narratives such as clients’ records and stories. For example, it is important in mental health care industry that most of the cases in mental health care are retained in terms of narratives. Narratives provide information how people deal successfully or unsuccessfully with real life problems. People can have a more comprehensible understanding on their difficulties and challenges by listening to the other’s similar stories since they are easy to remember, easy to understand and deal with human-like experiences (Gabriel, 2000). These stories also help people to adapt to the experience and discover new innovative ideas from others in order to solve their own problems (Lämsä & Sintonen, 2006).

In this paper, a narrative-based reasoning (NBR) algorithm is proposed with applications in decision support in mental health care. The proposed method aims at automatically processing the unstructured narrative information and suggesting recommendations for problem solving. It integrates the technologies of knowledge-based system (KBS), natural language processing (NLP) and artificial intelligence (AI). The NBR algorithm does not only help the knowledge workers to search relevant information easier by the enhanced indexing capability but also provides useful and reliable recommendations for the workers in problem solving so as to increase the quality of their solutions. A series of experiments have been carried out for measuring the performance of the purposed method based on real cases conducted in a Hong Kong based social service organization.