Methods for producing summaries from structured data have gained interest due to the huge volume of available data in the Web. Simultaneously, there have been advances in natural language generation from Resource Description Framework (RDF) data. However, no efforts have been made to generate natural language summaries for groups of multiple RDF entities. This paper describes the first algorithm for summarising the information of a set of RDF entities in the form of human-readable text. The paper also proposes an experimental design for the evaluation of the summaries in a human task context. Experiments were carried out comparing machine-made summaries and summaries written by humans, with and without the help of machine-made summaries. We develop criteria for evaluating the content and text quality of summaries of both types, as well as a function measuring the agreement between machine-made and human-written summaries. The experiments indicated that machine-made natural language summaries can substantially help humans in writing their own textual descriptions of entity sets within a limited time.

Clustering accuracy (ACC) is one of the most often used measures in literature to evaluate clustering quality. However, the measure is often used without any definition or reference to such a definition. In this paper, we identify the origin of the measure. We give a proper definition for the measure and provide a simple bug fix which allows it to be used also in the case of a mismatch in the number of clusters. We show that the measure belongs to a wider class of set-matching based measures. We compare its properties to centroid index (CI) and normalized mutual information (NMI).

Memory disorders and dementia are a central factor in the decline of functioning and daily activities in older individuals. The workload related to standardized speech tests in clinical settings has led to a growing emphasis on developing automatic machine learning techniques for analyzing naturally spoken speech. This study presented a bag of acoustic words approach for distinguishing dementia patients from control individuals based on audio speech recordings. In this approach, each individual’s speech was segmented into voiced periods, and these segments were characterized by acoustic features using the open-source openSMILE library. Word histogram representations were formed from the characterized speech segments of each speaker, which were used for classifying subjects. The formation of word histograms involved a clustering phase where feature vectors were quantized. It is well-known that partitional clustering involves instability in clustering results due to the selection of starting points, which can cause variability in classification outcomes. This study aimed to address instability by utilizing robust K-spatial-medians clustering, efficient K-means++ clustering initialization, and selecting the smallest clustering error from repeated clusterings. Additionally, the study employed feature selection based on the Wilcoxon signed-rank test to achieve computational efficiency in the methods. The results showed that it is possible to achieve a consistent 75% classification accuracy using only twenty-five features, both with the external ADReSS 2020 test data and through leave-one-subject-out cross-validation of the entire dataset. The results rank at the top compared to international research, where the same dataset and only acoustic features have been used to diagnose patients.

The goal of this work was to experiment with the formal modeling and automated reasoning of concurrent/parallel systems, mainly focusing on mutual exclusion algorithms. A concrete method is presented, which depends on timed automata and the model checker provided by the popular Uppaal toolbox. The method can be used for a thorough assessment of the properties of mutual exclusion algorithms. The paper argues that the proposed approach can be effective in moving beyond informal and intuitive reasoning about concurrency, which can be challenging due to the non-determinism and interleaving of the action execution order of the involved processes. The approach is described and applied to an in-depth analysis of Peterson’s algorithms for N= 2 and N> 2 processes. The analysis allows the reconciliation of different evaluations reported in the literature, particularly for the overtaking bound, and also reveals new results not previously disclosed. The general case of N> 2 was handled within the context of the state-of-art, standard, and efficient tournament binary-tree organization, which uses the solution for two processes to arbitrate between pairs of processes at different stages of their upward movement along the tree. All mutual exclusion algorithms are investigated under both atomic and non-atomic memory models.

In this paper, the classical coot optimization algorithm (COA) is modified to improve its overall performance in the exploration phase by adding an adaptive sigmoid inertia weight-based method. The modified coot optimization algorithm (mCOA) was successfully assessed using 13 standard benchmark test functions, which are frequently used to evaluate metaheuristic optimization algorithms. The MATLAB software was utilized to conduct simulation tests, and the outcome was compared with the performance of the original COA, the particle swarm optimization, and the genetic algorithm reported in the literature. The findings showed that the proposed algorithm outperformed the other algorithms on ten (10) of the 13 benchmark functions, while it maintained a competitive performance on the remaining three benchmark test functions. This indicates that mCOA provides a significant improvement to the original COA, thus making it suitable for resolving optimization problems in diverse fields. As a result, the proposed algorithm is recommended for adoption to solve real-life engineering optimization problems.