The aim of the article is analysis using LOF, COF and Kmeans algorithms for outlier detection in rule based knowledge bases. subject mining very important nowadays. Outliers rules mean unusual which are rare comparison to others should be explored further by domain expert. In research authors use methods find a given (1%, 5%, 10%) number outliers rules. Then, they analyze seven various quality indices, that used all after removing selected outliers, improve clusters. experimental stage six...

In this article, we evaluate the efficiency and performance of two clustering algorithms: AHC (Agglomerative Hierarchical Clustering) K-Means. We are aware that there various linkage options distance measures influence results. assess quality using Davies-Bouldin Dunn cluster validity indexes. The main contribution research is to verify whether clusters without outliers higher than those with in data. To do this, compare analyze outlier detection algorithms depending on applied algorithm....

Rule-based knowledge bases are constantly increasing in volume, thus the stored as a set of rules is getting progressively more complex and when not organized into any structure, system inefficient. The aim this paper to improve performance mining b ases by modification both their structure inference algorithms, which author's opinion, lead efficiency process. good approach shown through an extensive experimental study carried out on collection real bases. Experiments prove that partition...

Earlier studies have indicated that consumption of beavers Castor ssp. by wolves Canis lupus varies seasonally and is influenced rainfall affecting water levels. Therefore, to determine whether these carnivores prey more often on in drier seasons years, we assessed the diet Wigry National Park (NE Poland) analysing 303 scats collected from 2017 2019. The most important this region was wild boar Sus scrofa (25.2% consumed biomass), Eurasian beaver fiber (24.4%) roe deer Capreolus capreolus...

Decision support systems founded on rule‐based knowledge representation should be equipped with rule management mechanisms. Effective exploration of new in every domain human life requires algorithms organization and a thorough search the created data structures. In this work, author introduces an optimization both base structure inference algorithm. Hence, new, hierarchically organized is proposed as it draws cluster analysis method forward‐chaining algorithm which searches only so‐called...

Detecting outliers is a widely studied problem in many disciplines, including statistics, data mining, and machine learning. All anomaly detection activities are aimed at identifying cases of unusual behavior compared to most observations. There methods deal with this issue, which applicable depending on the size set, way it stored, type attributes their values. Most them focus traditional datasets large number quantitative attributes. The multitude solutions related detecting sets, still...

Knowledge engineering and data mining are the two biggest pillars of modern intelligent systems [...]

In this work the topic of applying clustering as a knowledge extraction method from real-world data is discussed. Authors propose two-phase cluster creation and visualization technique, which combines hierarchical density-based algorithms1. What more, authors analyze impact sampling on result searching through such structure. Particular attention was also given to problem visualization. review selected, two-dimensional approaches, stating their advantages drawbacks in context representing...

The article concerns the detection of outliers in rule-based knowledge bases containing data on Covid 19 cases. authors move from automatic generation a base source by clustering rules to optimize inference processes and detecting unusual allowing for optimal structure rule groups. paper presents two-phase procedure, wherein first phase, we look clusters when there are outlier base. In second detect using LOF (Local Outlier Factor) algorithm. Then eliminate database check whether selected...

The aim of the study was to determine relationship between maternal age at delivery and selected properties cord blood stem cells. included 50 pregnant women aged 18 38 years in which spontaneous labors or elective cesarean sections were performed. Umbilical collected immediately after delivered newborns. samples analyzed Polish Stem Cells Bank Warsaw. highest mean WBC level (p < 0.05) observed umbilical from patients 35 more. Similarly, cell viability There no statistically significant...

The article presents both methods of clustering and outlier detection in complex data, such as rule-based knowledge bases. What distinguishes this work from others is, first, the application algorithms to rules domain bases, secondly, use detect unusual aim paper is analysis using four for bases: Local Outlier Factor (LOF), Connectivity-based (COF), K-MEANS, SMALLCLUSTERS. subject mining very important nowadays. Outliers If-Then mean rules, which are rare comparing should be explored by...

Inteligentne systemy wspomagania decyzji to interaktywne programy komputerowe, których celem jest zbieranie, przetwarzanie i analizowanie ogromnych ilości danych koniecznych do wskazania najlepszego rozwiązania. Na podstawie wiedzy zdobytej od ekspertów dziedzinowych inteligentny system ekspertowy w oparciu o wbudowane algorytmy wnioskowania symulujące myślenie ludzkie przetwarza posiadaną wiedzę (zgromadzoną systemie pozyskaną użytkownika, który konsultuje się z takim systemem), by...