Marcel Wehrle

Striving for semantic convergence with fuzzy cognitive maps and graph databases

The exponentially rising amounts of urban data demand new conceptual and technical methods for their management and storage. The era of the Semantic Web requires a convergence toward commonly shared meanings. The combination of fuzzy cognitive maps with graph databases is a first approach to a solution. This article determines the basic requirements for the storing of fuzzy cognitive maps to test current graph databases for their structural suitability. Six out of 47 graph databases fulfill the requirements and are thus recommended for further research purposes. As a proof of concept, OrientDB is used to present how a semantic convergence can be reached through the combination of fuzzy cognitive maps and graph databases in a cognitive city by tackling fuzziness.

Transforming fuzzy graphs into linguistic variables

Fuzzy graphs (FG) are capable of showing dependencies and relationships between each other to a certain degree. Often, these relationships are described by numbers, which impedes interpretability for humans because they communicate using natural language. This paper seeks to turn the mathematical output of an FG into natural language sentences by applying Restriction-Centered Theory (RCT) to enhance the possibilities of knowledge transfer for humans via an FG. The proposed framework connects FGs and the RCT to produce not only verbalized dependencies but also statements about the dependencies of FGs. As a proof of concept, a use case is introduced, where Swiss Airlines connecting passenger flows are analyzed. The statements of the frameworks output are verified by an expert at the company that owns the data.

Building a granular knowledge monitor: An application of the A.U.R.A framework

Gaining the ability to identify, assess and display what knowledge a user holds, especially in a digital environment that is designed to facilitate its exchange, is of great value. In this paper the authors will present a newly designed framework for this task and a prototype to demonstrate its applicability. The introduced framework will be able to cope with different data types, as well as explicitly and implicitly provided information. Furthermore, is being ensured that the various characteristics of knowledge, such as vagueness, imprecision and abstraction are accounted for. The developed prototype is presented in the final section of this paper.

Possible application domains for the knowledge carrier finder system

The ability to share knowledge among members of a community, has always been of great importance. Web-based technologies have facilitated this endeavour by offering tools that are specifically designed for this task. Throughout this paper, the authors will present a novel type of platform, which is designed to support users in their quest to share knowledge. It consists of a question and answer platform, with a built in recommendation system for knowledge carriers. The hybrid design ensures that questions be answered in a timely manner and by most suitable users. The developed prototype and an evaluation of application domains are presented in the final section of this paper.

Verbalization of Dependencies Between Concepts Built Through Fuzzy Cognitive Maps

The new computing paradigm known as cognitive computing attempts to imitate the human capabilities of learning, problem solving, and considering things in context. To do so, an application (a cognitive system) must learn from its environment (e.g., by interacting with various interfaces). These interfaces can run the gamut from sensors to humans to databases. Accessing data through such interfaces allows the system to conduct cognitive tasks that can support humans in decision-making or problem-solving processes. Cognitive systems can be integrated into various domains (e.g., medicine or insurance). For example, a cognitive system in cities can collect data, can learn from various data sources and can then attempt to connect these sources to provide real time optimizations of subsystems within the city (e.g., the transportation system). In this study, we provide a methodology for integrating a cognitive system that allows data to be verbalized, making the causalities and hypotheses generated from the cognitive system more understandable to humans. We abstract a city subsystem—passenger flow for a taxi company—by applying fuzzy cognitive maps (FCMs). FCMs can be used as a mathematical tool for modeling complex systems built by directed graphs with concepts (e.g., policies, events, and/or domains) as nodes and causalities as edges. As a verbalization technique we introduce the restriction-centered theory of reasoning (RCT). RCT addresses the imprecision inherent in language by introducing restrictions. Using this underlying combinatorial design, our approach can handle large data sets from complex systems and make the output understandable to humans.

Granular Computing – Fallbeispiel Knowledge Carrier Finder System

Die automatisierte Strukturierung und Speicherung von komplexen, semistrukturierten Daten in einer graph-basierten Wissensdatenbank erfordert eine logische und auch hierarchische Verknupfung, damit die Information fur Anwender und Applikationen leichter zu erreichen und interpretieren sind. Der Aufbau ermoglicht Informationen granular zu clustern und in einer mehrschichtigen, hierarchischen Struktur abzubilden. Die Visualisierung so entstandenen Granules, erlaubt die Entdeckung von Wissen aus bestehenden Informationen. Im Fallbei-spiel werden webbasierte, semistrukturierte Daten einer Question and Answer Plattform in einer Graph Datenbank abgebildet, visualisiert und als Teil eines Knowledge Carrier Finder Systems beschrieben.

A granular approach for identifying user knowledge

The ability to accurately assess the type and extent of knowledge a user possesses without having to demand that it is explicitly declared is of importance for a range of different applications. To obtain this ability, the authors have developed an approach that harnesses the benefits of granular computing in order to apply granular and hierarchical structure to knowledge. In this way, identification of different knowledge domains and degrees of granularity becomes possible that is essential when assessing a users knowledge. Furthermore, a method is described that could be used to improve the identification of knowledge by not relying exclusively on what a user has contributed. Content that is related to previous contributions and matches certain criteria can also be included in a bid to obtain accurate and richer predications of what knowledge a user possesses.