Search

The main aim of the paper is to discuss the architecture for the future Intelligent Systems (IS's) and Decision Support Systems (DS's) dealing with complex phenomena such as supporting medical decisions (diagnosis and therapy) and to emphasize challenges in designing such systems. More precisely, the paper presents arguments for developing a specialized computing model based on the interactive granular computing paradigm which can help to design IS's and DS's more close to the prototypes of real life decision making. In this regard, the paper brings to the fore different experiences faced during designing other medical IS's or DS's.As a starting step, the paper considers the experience of developing the OvuFriend platform and outlines some possible extension of it in the framework of the proposed architecture on the basis of Interactive Granular Computing (IGrC) model. Specifically, our attempt is to analyze a scheme, which is being used in the platform of OvuFriend for determining health risks and possibilities of a woman to conceive a child, from the perspective of IGrC. The target of the paper is two fold. Firstly, to show how the underlying AI algorithm of this scheme can be related with the notion of computing in the context of IGrC. Secondly, to identify possible extensions of the existing scheme so that it becomes more dynamic, interactive, and close to personalized medicine. [ABSTRACT FROM AUTHOR]

The neighborhood rough set model serves as an important tool for handling attribute reduction tasks involving heterogeneous attributes. However, measuring the relationship between conditional attributes and decision in the neighborhood rough set model is a crucial issue. Most studies have utilized neighborhood information entropy to measure the relationship between attributes. When using neighborhood conditional information entropy to measure the relationships between the decision and conditional attributes, it lacks monotonicity, consequently affecting the rationality of the final attribute reduction subset. In this paper, we introduce the concept of neighborhood granularity and propose a new form of relative neighborhood granularity to measure the relationship between the decision and conditional attributes, which exhibits monotonicity. Moreover, our approach for measuring neighborhood granularity avoids the logarithmic function computation involved in neighborhood information entropy. Finally, we conduct comparative experiments on 12 datasets using two classifiers to compare the results of attribute reduction with six other attribute reduction algorithms. The comparison demonstrates the advantages of our measurement approach. [ABSTRACT FROM AUTHOR]

Few-shot learning, which aims to identify new classes with few samples, is an increasingly popular and crucial research topic in the machine learning. Recently, the development of deep learning has deepened the network structure of a few-shot model, thereby obtaining deeper features from the samples. This trend led to an increasing number of few-shot learning models pursuing more complex structures and deeper features. However, discarding shallow features and blindly pursuing the depth of sample feature levels is not reasonable. The features at different levels of the sample have different information and characteristics. In this paper, we propose a few-shot image classification model based on deep and shallow feature fusion and a coarse-grained relationship score network (HFFCR). First, we utilize networks with different depth structures as feature extractors and then fuse the two kinds of sample features. The fused sample features collect sample information at different levels. Second, we condense the fused features into a coarse-grained prototype point. Prototype points can better represent the information in this class and improve classification efficiency. Finally, we construct a relationship score network, concatenating the prototype points and query samples into a feature map and sending it into the network to calculate the relationship score. The classification criteria for learnable relationship scores reflect the information difference between the two samples. Experiments on three datasets show that HFFCR has advanced performance. [ABSTRACT FROM AUTHOR]

In the era of big data, handling massive datasets to extract valuable information has become increasingly critical. Knowledge representation emerges as a pivotal method to address this challenge. In the domain of knowledge representation, there exist two primary approaches: symbolic representation and vector representation. The integration of symbolic and vector representations to harness their respective strengths has become the cutting-edge approach to address challenges in the field of knowledge representation. This paper proposes a method that integrates a partial order formal structure analysis (POFSA) with graph representation learning. Specifically, we initially construct three-way partial order structure graphs, then create an attribute graph based on this structure, which can be processed by the graph representation learning methods. Finally, we utilize the graph representation learning to construct embeddings for three-way attribute partial order structure diagram (APOSD). We comprehensively assesse these embeddings across eight different datasets and present the results. The experiments indicate the feasibility of our proposed approach which is proven a novel approach of combining symbolic and vector representations for handling complex data and implicit knowledge. [ABSTRACT FROM AUTHOR]

Density-based clustering algorithms are known for their ability to detect irregular clusters, but they have limitations when it comes to dealing with clusters of varying densities. In this paper, we propose a new clustering algorithm called granule fusion density-based clustering with evidential reasoning (GFDC). The approach introduces the concept of sparse degree, which measures both the local density and global density of samples. The sparse degree of samples reflects the stability of samples. Moreover, a core-granule is composed of the neighborhood granule of a sample, of which the sparse degree is minimum in its neighborhood. Then, the core-granules are generated based on the sparse degree and are insensitive to clusters with varying densities. The core samples, which consist of samples in core-granules, are used to form initial clusters through fusion strategies. Additionally, an assignment method is developed from Dempster-Shafer theory to assign border samples and identify outliers. The experimental results demonstrate the effectiveness of GFDC on extensive synthetic and real-world datasets. [ABSTRACT FROM AUTHOR]

By effectively capturing the similarity relationship between objects, neighborhood rough set models offer a highly advantageous approach to analyzing numerical data. However, it is computationally demanding to construct granular structures by calculating the distance between all objects in the universe. In this paper, we propose a novel approach to discretizing information systems associated with neighborhood rough sets. Firstly, a unique bilayer, quasi discretization-based neighborhood, is developed with the aim of minimizing the time required for model generation. Most of the current evaluation indices for a rough set model primarily focus on the evaluation of the lower and upper approximations, rather than the granular structure of the model. We introduce two evaluation indices for binary relations, namely, the Gaussian balance index and the quality index. These indices aim to provide a comprehensive assessment of the granular structure in terms of quantity and quality. Moreover, we present two extension models for discretization neighborhood rough sets and devise three attribute reduction algorithms. In order to evaluate the performance of the models, three comparative experiments are conducted. The results obviously reveal the high efficiency and effectiveness of the proposed models. [ABSTRACT FROM AUTHOR]