2,979 publications from this institution
This paper presents a comparative analysis of Deep Learning models and Fuzzy Rule-Based Classifiers (FBRCs) for Brain Tumor Classification from MRI images. The study considers a publicly available dataset with three types of brain tumors and evaluates the models based on their accuracy and complexity. The study involves VGG16, a convolutional network known for its high accuracy, and FBRCs generated via a multi-objective evolutionary learning scheme based on the PAES-RCS algorithm. Results show that VGG16 achieves the highest classification performance but suffers from overfitting and lacks interpretability, making it less suitable for clinical applications. In contrast, FBRCs, offer a good balance between accuracy and explain- ability. Thanks to their straightforward structure, FRBCs provide reliable predictions with comprehensible linguistic rules, essential for medical decision-making.
International audience
This paper presents a digital, transistor level implemented neo-fuzzy neural network. This type of neural network is particularly well suited for real-time applications like those encountered in signal processing and nonlinear system identification. We consider in detail a flexible reconfigurable circuit of a single nonlinear synapse of this network. When combining such circuits, single-layer or multilayer networks can be designed. The advantages of the proposed circuit come in the form of reduced redundancy, high data rate due to parallel operation, low power consumption, and an overall flexibility of system configuration.
Solving optimization problems under hybrid uncertainty bears a heavy computational burden. In this study, we propose a unified structured optimization approach, termed robust granular optimization (RGO), to tackle the optimization problems under hybrid manifold uncertainties in a computationally tractable manner. Essentially, the RGO can be regarded as a complementary fusion of granular computing and robust optimization techniques. The paradigm of RGO consists of three core phases: 1) uncertainty identification, 2) information granulation in which basic granular units (BGUs) are formed, and 3) robust optimization realized over the BGUs. Following the proposed paradigm, we develop two classes of RGO models for general single-stage and two-stage optimization problems with separable and higher order hybrid uncertainties, respectively. It is shown that both types RGO models can be equivalently transformed into linear programs or mixed integer linear programs that can be handled efficiently by off-the-shelf solvers. Furthermore, a target-based tradeoff model is developed to enhance the flexibility of the RGO models in balancing the granularity level (or robustness level) and the solution conservativeness. The tradeoff model can also be efficiently solved by a binary search algorithm. Finally, sufficient computational studies are presented, and comparisons with the existing approaches show that the RGO models can bring much higher computational efficiency and scalability without losing much optimality, and the RGO solutions exhibit a stronger resistance to the uncertainty.
In this study, we discuss the problem of interpretability of scale versus polarity in multicriteria decision-making problem. Decision making requires aggregation of premises of different characters and types. The influence of premises on a decision to be made may have a different characteristics as well. Some premises may have a positive character, i.e. they vote/agitate towards making a decision, others may fight against a decision. On the other hand, premises can be tempered by priorities, which may affect their character. Therefore, there is a need to discuss different configurations of premises and their priorities. This is the first aspect of our discussion on multicriteria decision making. The second one under discussion is the interpretability of all aspects mentioned so far. In this case, we discuss the representation problems of both premises and priorities. They are usually exhibited as numbers taken from some scale as, for instance, the unipolar unit interval [0,1] or the bipolar unit interval [-1,1]. On the other hand, there is a question raised about a character of premises/priorities, that is, whether they vote pro or contra a decision to be taken and what relations between scales and polarities are.
Abstract This overview is focused on the book reflecting research results on the fundamentals of the theory of multicriteria (multiobjective and multiattribute) decision-making under conditions of uncertainty. The facet of uncertainty is formalized based on a possibilistic (not probabilistic) approach. These results are based on the fuzzy set theory and its fusion with other branches of mathematics of uncertainty. The overview identifies the crucial arguments behind the ultimate need for this theory, reflects the book’s primary objectives, identifies the key possibilities delivered by the presented book's results, and elaborates on real-world problems solved by applying the findings reported in the book. The thorough critical analysis summarizes the advantages and limitations of the main results covered by the book.
Concept formation contrives one among vital issues in all fields of science. It can be stated that, to a significant degree, a scientific discovery is related to various aspects of creation of general categories out of a mass of raw empirical data analyzed from a suitable perspective. Thus, as it is seen now, the concept of any notion is constructed on the basis of a significant amount of previous experience. Despite a lot of research completed, there still exists a number of open questions about diverse features of the constructed concepts. In this article we develop a framework for concept formation making use of a logical platform of fuzzy sets combined with mechanisms of neurocom-putations. It will be indicated how for a given naming of the concept its description can be derived. Moreover, a particular attention will be devoted to mechanisms justifying a relevance of the concepts with respect to a collection of available empirical facts. This may play a primordial role in recognizing some limitations of a character of the concept (i.e., its generality or specificity) which cannot be exceeded simultaneously not losing consistency with the family of collected objects utilized within the process of concept formation.
Abstract Transcriptional regulation mainly controls how genes are expressed and how cells behave based on the transcription factor (TF) proteins that bind upstream of the transcription start sites (TSSs) of genes. These TF DNA binding sites (TFBSs) are usually short (5-15 base pairs) and degenerate (some positions can have multiple possible alternatives). Traditionally, computational methods scan DNA sequences using the position weight matrix (PWM) of a given TF, calculate binding scores for each K-mer against the PWM, and finally classify a K-mer as to whether it is a putative TFBS or a background sequence based on a cut-off threshold. The FSCAN system, which is proposed in this paper, employs machine learning techniques to build a learner model that is able to identify TFBSs in a set of bound sequences without the need for a cut-off threshold. Our proposed method utilizes fuzzy inference techniques along with a distribution-based filtering algorithm to predict the binding sites of a TF given its PWM model and phastCons scores for the input DNA sequences. Data imbalance reduction techniques are also used to ease the learning of the adaptive-neuro fuzzy inference system (ANFIS) algorithm. The proposed system is tested on 22 ChIP-chip sequence-sets from the Saccharomyces Cerevisiae genome. Our results show that FSCAN outperforms other approaches like MatInspector and MATCH and is quite robust. As more transcriptional data becomes available, our proposed framework encourages the use of fuzzy logic techniques in the prediction of TFBSs.
