روش ساخت و ساز روایی محاسباتی با برنامه های کاربردی در یادگیری سازمانی از سازمان های خدمات اجتماعی

Acquisition of knowledge must be interwoven with the process of applying it. However, traditional training methods which provide abstract knowledge have shown ineffective for gaining experience of the work. In order to solve this problem, more and more researchers have included narrative in simulation, which is known as narrative simulation. By providing the narratives, participants recognize the choices, decisions, and experience that lead to the consequences of those decisions. It has been proven that narrative simulation is very useful in facilitating in-depth learning and reflective learning. However, conventional methods of data collection and narrative construction for narrative simulation are labor intensive and time consuming. They make use of previous narratives manually and directly. They are inadequate to cope with the fast moving world where knowledge is changing rapidly. In order to provide a way for facilitating the construction of narrative simulation, a novel computational narrative construction method is proposed. By incorporating technologies of knowledge-based system (KBS), computational linguistics, and artificial intelligence (AI), the proposed method provides an efficient and effective way for collecting narratives and automating the construction of narratives. The method converts the unstructured narratives into a structural representation for abstraction and facilitating computing processing. Moreover, it constructs the narratives that combine multiple narratives into a single narrative by applying a forecasting algorithm. The proposed method was successfully implemented in early intervention in mental health care of a social service company in Hong Kong since the case records in that process have structural similarities to narrative. The accuracies of data conversion and predictive function were measured based on recall and precision and encouraging results were obtained. High recall and precision are achieved in the data conversion function, and high recall for the predictive function when new concepts are excluded. The results show that it is possible for converting multiple narratives into a single narrative automatically. Based on the approach, it helps to stimulate knowledge workers to explore new problem solving methods so as to increase the quality of their solutions.

Mental health problems impact seriously to the society. The impact of mental disorders can be reduced by identifying the distress at an early stage, establishing an early and accurate diagnosis and providing prompt and effective treatment. This idea underlies the interest in early intervention in mental disorders. Some countries have put this as a major element in their mental health policy (Kemp, 1993). Evidence from many evaluation studies suggests that well-designed and intensive early intervention programs have the potential to yield outcomes that benefit health plans (e.g., improved health outcomes, lower health care costs, lower maternity costs, fewer emergency room visits) and the outcomes that have potential benefits for Medicaid, the government, and society as a whole such as higher educational attainment, greater economic self-sufficiency, lower crime rate, etc. (Perloff, Butler, Berry, & Budetti, 1998). However, there is a variety of challenges faced by the mental health social service providers (Ferns, 1995). The increased need for services, decreased subvention for services due to the economic restructuring and the attendant quest for budget cuts, and growing government regulation lead to the formulation of an immense pressure to social service organization to provide effective, customized and high-quality care at the lowest cost and greater administrative control (Savage, 1987). The social service providers are facing the problem of conflict between these objectives. Limited resources must be traded off in order to accomplish any one of the objectives over the others. Shrinking revenues have forced the social service providers to look for creative ways to provide quality services at less expense. As a result, any techniques or methods that sustain knowledge growth and distribution are keys to quality services (Von Krogh & Grand, 1999). The most typical way of learning is providing training for the workers. In the traditional model of on-the-job training, workers would typically receive a pre-prepared course in the new regulations, procedures, or processes to promote the new practices. They are often at a different location from their place of work and they are expected to apply this abstracted knowledge later in their workplace. As indicated by the figures of training effectiveness, the results are often observed not to be encouraging. Businesses spend up to USD$100 billion per year to train up workers. Yet estimates are that less than 10% of this training transfers to the job (Detterman, 1993). Norman, 2000 and Norman, 2003 also presents a very interesting discussion of full-curriculum interventions versus small-scale laboratory studies and concludes that curriculum-wide studies are not worth the effort involved in doing them. Current theories of learning reveal that the knowledge acquisition process cannot be separated from the process of applying it. The effective integration of working and learning is a fundamental requirement for businesses to remain competitive. Brown and Duguid (1991) also argue that learning is the essential bridge between working and innovation, and the three processes are inextricably intertwined. They argue that on-the-job training separates simplified abstract principles from the rich detail of actual practice and separates learners from the workplace community. Instead, they advocate that technology and business processes should support the existing rich learning practices within the workplace community by enabling individuals within communities to somehow retain and share their experience. This influential article has inspired many projects in knowledge management (KM) and organizational learning (OL). Hence, the approach of acquisition and sharing existing experience is the key goal of many OL approaches (Landes, Schneider, & Houdek, 1998). Numerous learning systems have been developed for the retention of information and knowledge of organizational problems. Some of them are dedicated software tools intended to offer a virtual educational and/or online training environment. They provide dynamic capabilities to acquire and share the knowledge within an organization so that the knowledge workers can learn and cope with their work effectively. Despite a large number of functions covering a large number of users’ needs, the traditional methodologies of learning systems are fundamentally limited. These tools provide past information or cases for training and decision support, which is insufficient to cope with the complex, diverse and continuously evolving business environment. Such kind of systems may hinder innovation by providing previous experience for problem solving. Recently, more and more researchers have incorporated narrative into learning. Such kind of manual simulation practice is named narrative simulation. It is interesting to note that narrative simulation is very useful in facilitating in-depth learning and reflective learning (Cole, 1997). According to Argyris (1977), reflective learning is the core of organizational learning. Reflection can be seen as “consciously thinking about and analyzing what one has done (or is doing)”. Reflective Learning is a structural approach that enables learners to reflect upon their learning, to understand their own learning processes and thus allow them to become more autonomous. However, traditional narrative simulation makes use of previous narratives manually and directly which is inadequate to cope with the fast moving world where knowledge within organizations is changing rapidly and continuously updating is vital. Since the case records in early intervention in mental health care have structural similarities to narrative, a computational narrative construction approach is proposed in this paper for providing a reflective training of early intervention in mental health care. The proposed approach aims at collecting the narratives and constructing a scenario by combining the collected narratives automatically. It integrates the technologies of knowledge-based system (KBS), natural language processing (NLP), artificial intelligence (AI) and computational forecasting method. Based on the approach, it provides extra time for the knowledge workers for exploring new problem solving methods so as to increase the quality of their solutions. A series of experiments based on real cases have been carried out for measuring the accuracy of data conversion and the accuracy of narrative prediction.

Traditional training methods which provide abstracted knowledge are inadequate for sustaining knowledge growth. For people learn through actions, it is important for people to be exposed to the problems so that they can gain more experience. However, wrong decisions are painful and costly. Narrative simulation provides a solution which allows people to try their actions in a way that is cost effective, faster, appropriate, flexible, and ethical. It is very useful in facilitating in-depth learning and reflective learning which is the core of organizational learning. The application of narrative simulation increases the impact of the experience for users to rethink their daily work. This is particularly important because most of the valuable new knowledge is generated in this process. However, conventional methods make use of previous narratives manually and directly are inadequate to cope with this fast moving world. In this paper, a computational narrative construction method, which incorporates technologies of knowledge-based system (KBS), computational linguistics, and artificial intelligence (AI), is proposed for automating the narrative construction methods. It overcomes the limitation of traditional methods which collect and construct narratives manually, it converts unstructured narrative information into a structural format, and it automatically combines multiple narratives into a single narrative. To verify the performance of the proposed method, real case experiments were carried out in mental health care for professional development. The experiments measure the accuracy of conversion from unstructured text into structured concept maps and the accuracy of the predictive function. The results show that there is a high recall (0.93%) and a high precision (0.90%) for the data conversion function. In the prediction of concepts and propositions, relative high recall rates are achieved (Over 0.65 for prediction of concepts and over 0.65 propositions when the concepts and propositions that appeared at the first time are excluded and the time interval is set to 3 months or above.) However, low precision rate (less than 0.15) is resulted. Further work will be done on measuring the similarity among the propositions so as to minimize the number of propositions and improve the performance of the method. In addition to the social service and health industry, potential applications of the tool can also be found in other industries.