We propose a generic model of an intelligent collective that emulates social communication. Its main elements are interchangeable, which allows fast building of specific collective models. The key point is the asynchronous communication between agents. The model allows changes in knowledge representation, modes of communication, algorithms used to receive new knowledge and memory algorithms dealing with forgetting older knowledge. We discuss multiple variants of all those elements, including communication based on Twitter and Facebook social networks. We test examples of specific collectives built using the generic model.

In this paper, we consider the known opinion formation problem from the point of view of computational collective intelligence research, with consideration to some sociological research in the area of social influence. We study two approaches to opinion formation – polarization and persuasion – as viewed from the point of view of our collective model (as opinion integration strategies). We detail both approaches and then conduct simulational experiments to study them both in single and averaged over multiple runs. We found special interest in the polarization strategy, which we want to use in the future as a basis for a new measure of collectives behavior.

Knowledge diffusion and opinion formation are important research topics in the area of communication in social networks. We present our generic model with interchangeable variants for processing messages from other members of the group. We describe several generic aspects of communication, like learning, persuasion, or polarization, and apply them in our model. We then show how this simple change leads to different results for the communication process of the whole group. We show how the model with interchangeable integration strategies is applicable for modeling various aspects of online social network and real life communication.

We present a model of a group of identical agents that may be used to simulate knowledge diffusion in social networks. At this time the model is still in development phase, but it includes notions derived from sociological theories as well as mathematical apparatus from the collective intelligence area. In this paper the model is extended by dynamic memory of agents, allowing them to forget knowledge. The model is tested in simulations based on criteria that were derived from sociological research describing functional real-world groups. We discuss in detail the combination of model elements giving most realistic results and present description of additional details that need to be included in the model to make it more similar to real-world groups.

With the increasing availability and production of data, data-driven decision-making has been largely adopted by people and organizations. However, finding and purchasing data suitable for a specific purpose and addressing unique (and sometimes) personal requirements is still a challenge. Alternative solutions to the traditional data commerce, the data marketplaces, are of almost no help for novice buyers, as the number of datasets that are available to purchase is often massive. To alleviate this problem, we consider an intelligent decision support approach for helping buyers, particularly inexperienced ones, to better search and evaluate data offerings. The contribution of this work consists of a preference recommendation strategy that focuses a buyer's attention towards more appropriate offers while enhancing his or her awareness of the market. An experimental case is designed to show the feasibility and effectiveness of the approach. The results obtained demonstrate that preference recommendation has the potential to suggest unexpected, yet useful data offerings to buyers.

One of basic divisions of information retrieval systems is content-based and collaborative filtering. Some hybrid methods exist combining both of them, but certain aspects still remain unexplored. In this paper we explore one: the influence of users communicating via social media on content-based recommendation systems. While in the system users do not know each other, outside they may make their own preferences known (e.g. tweeting recommendations), thus influencing the preferences of other users. Here we simulate several different types of such communication and its influence on content-based recommendation system. We intend to use this results for improving the quality of such systems.

Knowledge integration is used more and more often in different theoretical and practical works. It is usually understood as an instantaneous change of knowledge states. In our research we consider integration as a process and study its different elements. Here we focus on using asynchronous communication between agents in a decentralized system. We add a structure of preferred connections and simulate different types of groups. We show that this approach works similar to the classical centralized one, while removing the need for a single critical agent. We also use this approach in a prototype of a weather prediction system to show that the results are similar to obtained with other modes of agent communication.

This paper presents issues related to a process of the automatic summarization of the text documents connected with economic knowledge performed by the cognitive agents in an integrated management information system.In contemporary companies, the unstructured knowledge is essential, mainly due to the possibility of obtaining better flexibility and competitiveness of the organization.Therefore more often the decision are taken in the enterprises on the basis of the summaries.The first part of the paper shortly presents the state-of-the-art in the considered field; next, the summarization process in the Cognitive Integrated Management Information System is characterized; the case study related with the summaries generating agent is presented in the last part of this paper.

Data warehouse federations are a next step in modern information storage and processing systems. They are a system of classical data warehouses connected by an integrated schema, that allows access to all the data of the source warehouses. We propose a generic approach to create a data warehouse federation, providing methods for schema integration (in order to create the federation schema), query decomposition (in order to translate the query to the federation into queries to subsequent warehouses) and result integration (to select the answer that best suits the user information need).We test this approach by creating a basic implementation and present a working example.

The process of integrating data in a hierarchical structure is usually based on algorithms designed for one specific task; for example, for eXtendable Markup Language (XML) schema integration. In this article, a more general approach for integration of data in a hierarchical structure is proposed using a generalized tree structure called a complex tree. For this aim we define a set of criteria for integration. These criteria are completeness, precision, and optimality. A simple algorithm based on these criteria is presented and the advantages of the approach are discussed.