Zoo Dataset Research Articles

Self-supervised learning on MeerKAT wide-field continuum images

Context. Self-supervised learning (SSL) applied to natural images has demonstrated a remarkable ability to learn meaningful, low-dimension representations without labels, resulting in models that are adaptable to many different tasks. Until now, applications of SSL to astronomical images have been limited to Galaxy Zoo datasets, which require a significant amount of preprocessing to prepare sparse images centered on a single galaxy. With wide-field survey instruments at the forefront of the Square Kilometer Array (SKA) era, this approach to gathering training data is impractical. Aims. We demonstrate that continuum images from surveys such as the MeerKAT Galactic Cluster Legacy Survey (MGCLS) can be successfully used with SSL, without extracting single-galaxy cutouts. Methods. Using the SSL framework DINO, we experimented with various preprocessing steps, augmentations, and architectures to determine the optimal approach for this data. We trained both ResNet50 and Vision Transformer (ViT) backbones. Results. Our models match state-of-the-art results (trained on Radio Galaxy Zoo) for FRI/FRII morphology classification. Furthermore, they predict the number of compact sources via linear regression with much higher accuracy. Open-source foundation models trained on natural images such as DINOv2 also excel at simple FRI/FRII classification; the advantage of domain-specific backbones is much smaller models trained on far less data. Smaller models are more efficient to fine-tune, and doing so results in a similar performance between our models, the state-of-the-art, and open-source models on multi-class morphology classification. Conclusions. Using source-rich crops from wide-field images to train multi-purpose models is an easily scalable approach that significantly reduces data preparation time. For the tasks evaluated in this work, twenty thousand crops is sufficient training data for models that produce results similar to state-of-the-art. In the future, complex tasks like source detection and characterization, together with domain-specific tasks, ought to demonstrate the true advantages of training models with radio astronomy data over natural-image foundation models.

Hybrid ABC–K Means for Optimal Cluster Number Determination in Unlabeled Data

This study presents the ABC K Means GenData algorithm, an enhancement over traditional K Means clustering that integrates the Artificial Bee Colony (ABC) optimization approach. The ABC K Means GenData algorithm addresses the issue of local optima commonly encountered in standard K Means algorithms, offering improved exploration and exploitation strategies. By utilizing the dynamic roles of employed, onlooker, and scout bees, this approach effectively navigates the clustering space for categorical data. Performance evaluations across several datasets demonstrate the algorithm's superiority. For the Zoo dataset, ABC K Means GenData achieved high Accuracy (0.8399), Precision (0.8089), and Recall (0.7286), with consistent performance compared to K Means and Fuzzy K Means. Similar results were observed for the Breast Cancer dataset, where it matched the Accuracy and Precision of K Means and surpassed Fuzzy K Means in Precision and Recall. In the Soybean dataset, the algorithm also performed excellently, showing top scores in Accuracy, Precision, Recall, and Rand Index (RI), outperforming both K Means and Fuzzy K Means.. The comprehensive results indicate that ABC K Means GenData excels in clustering categorical data, providing robust and reliable performance. Future research will explore its application to mixed data types and social media datasets, aiming to further optimize clustering techniques. .

Educational Practice Based on the Integration of Support Vector Machines and English Curricula

At present, IT (Information Technology) with computer network as its core has begun to integrate into CET (College English Teaching) practice, which has brought unprecedented influence on ELT (English Language Teaching). This paper adopts the strategy of active learning to build SVM, overcomes the shortcomings of typical SVM, and comprehensively considers various actual situations to build an assessment model of college English classroom instructional quality. The experimental results show that the classification accuracy of this algorithm can reach 96.21% on Zoo data set and 94.36% on Iris data set. This study is of great theoretical and practical value for improving the level of college English classroom teaching in the IT ecological environment.

Statistical method for clustering high-dimensional data based on fuzzy mathematical modeling

Abstract In this paper, firstly, the fuzzy mathematical model and its types and algorithms of fuzzy numbers are studied. Then, the concept of fuzzy mathematics is added to cluster analysis, and fuzzy cluster analysis is carried out on samples or objects after data standardization and the construction of a similarity matrix. Then, the fuzzy C-mean FCM algorithm is proposed, and the FCM algorithm is improved by introducing the point density of data objects, the maximum minimum distance method of point density sampling, and the reduction of computation to improve the statistical function of clustering high-dimensional data. Finally, the algorithm of this paper is compared with other algorithms by AC, PR, RE, convergence speed, running time, DBI index and other indexes in order to analyze the advantages of the FCM algorithm in high dimensional data clustering statistics. The results show that the F value of the FCM algorithm in the Zoo dataset reaches 0.976, which improves 61% compared to FKM and nearly 12.6% compared to the IIFKM0 algorithm, with the best clustering effect. The DBI index of the FCM algorithm is the lowest in the four datasets, which is basically stabilized in the range of 0.5-0.6. This is better than other algorithms.

An Efficient Controller Placement Algorithm using Clustering in Software Defined Networks

Software-defined networking (SDN) is an emerging network paradigm that separates the control plane from data plane and provides programmable network management. The control plane within SDN is responsible for decision-making, while packet forwarding is handled by the data plane based on flow entries defined by the control plane. The placement of controllers is an important research issue that significantly impacts performance of SDN. In this work, we utilize clustering algorithms to group networks into multiple clusters and propose an algorithm for optimal controller placement within each cluster. Initially, using the silhouette score, we determine the optimal number of controllers for various topologies. Additionally, to enhance network performance, we employ the meeting point algorithm to calculate the optimal location for placing the controller within each cluster. We implement the proposed work in the Mininet emulator with POX as the SDN controller. Furthermore, we compare our proposed approach to existing work in terms of throughput, delay, and jitter using six topologies from the Internet Zoo dataset.

HCDC: A novel hierarchical clustering algorithm based on density-distance cores for data sets with varying density

Cluster analysis is a crucial data mining technology widely used in image segmentation, language processing, and pattern recognition. Most existing clustering algorithms cannot identify complex shapes in manifold data sets and data sets with varying-density distribution, especially when clusters with significant differences in density are close to each other. Hierarchical clustering algorithms can identify data sets of arbitrary shapes. However, hierarchical clustering algorithms not only cannot cluster datasets with significant density variations but also have a high time cost. So in this paper, we propose a novel hierarchical clustering algorithm based on density-distance cores, called HCDC. It first selects the density-distance representative points for each point from the set of candidate representative points. Then it selects density-distance cores from all density-distance representatives. And it replaces the whole data set with density-distance cores and uses a new distance between them to apply hierarchical clustering. To avoid the influence of noise points in the dataset when finding density-distance cores, we also propose the noise point detection method and verify the feasibility of this method. In this paper, we compare our proposed algorithm with existing classical and novel algorithms on synthetic and real datasets. Experiments show that our algorithm clusters better than existing algorithms on complex-shaped datasets and datasets with different densities. On datasets with sparse and dense clusters close to each other, the ARI score of HCDC is more than 0.1 higher than that of LDP-MST. In particular, on the grid dataset, HCDC’s ARI score is 0.997 higher than LDP-MST. On DS3 and DS8, HCDC’s ARI score is more than 0.14 higher than the second-best algorithm, RNN-DBSCAN. Moreover, on the zoo dataset, HCDC’s ARI score is 0.15 and 0.6 higher than RNN-DBSCAN and LDP-MST, respectively. On the olivetti face dataset, HCDC is the only algorithm with an NMI score above 0.9 on photo1 and photo2 datasets.

SVM Classifier: Identify Linear Separability Of NAND And NOR Logic Gates

Abstract-Artificial Neural Network plays vital role to resolve several real-life problems. Those problems would be impossible or difficult to solve by human or statistical principles. ANN produces better results on large data set as it has self-learning capabilities. The most frequently used family of ANN is pattern classification. Generally, patterns are obtained from real world and get together based on definite attributes in specific regions. Data is classified based on knowledge by recognizing patterns. This study underlines linear separability of Boolean logic gates according to classification. We considered Boolean functions NAND and NOR for classification task. Our problem statement related to this study is, “There is significant difference between the performance of classifier regarding NAND, NOR Boolean logic gates using Support Vector Machine”. Zoo data set is used to carry out the experiment. Accuracy score is measured using Support Vector Machine (SVM). For actual and predicted data set, classification report and confusion matrix is produced. For classification SVM is emerged as a promising technique. Usually, SVM gives remarkable performance as compare to other Machine Learning classifiers methods. As a result, SVM produces outstanding performance on classification problem. Keywords- Boolean Functions, NAND, NOR, Linear Separability, Support Vector Machine (SVM), Confusion Matrix.

Evaluating the Effect of Optimized Voting Using Hybrid Particle Swarm and Grey Wolf Algorithm on the Classification of the Zoo Dataset

When there are numerous possible solutions for a given class in a given problem, majority voting or plurality voting is typically employed. One common technique for improving classification accuracy is bagging, which involves training many classifiers on slightly different datasets and then voting on the combined results. In this research, we examine how alternative voting procedures affect the efficiency of two distinct classification algorithms applied to datasets of varying complexity. Despite the increased computing cost associated with determining preference order, the results show that the single transferable vote can be a suitable alternative to plurality voting.

DETERMINING LEARNING SUCCESS of kNN ALGORITHM on ZOO DATASET

People learn by examining, observing and researching their environment. They actually gains experience from what they have learned. By using the experience they have gained, they can adapt to the new situation they encounter and make decisions. People always make decisions by comparing their previous knowledge while describing objects and classifying them. Similarities and differences to previously learned objects are very effective in decision making. It has been shown in the studies that the experiential learning method can also be used on machines. Intelligent machines and devices that use machine learning methods in their structure are widely used in many areas. Machine learning can be performed using different algorithms. These algorithms use the attributes of the objects in the data set when making decisions. Similarities and differences in the attributes of objects are obtained by comparing them with previous experiences. As a result of the comparison, a decision is made and predictions are made about the classes of the objects. In this study, kNN machine learning algorithm, which is a supervised learning method, was used on the Zoo dataset. In this data set, there are attributes of common living things. By using these attributes, the classes of living things in the data set are determined. The “k” neighbor value and weight parameter selected in the kNN algorithm affect the learning success. In this study, the effect of two parameters used in the kNN algorithm on learning success is shown. According to the results obtained, the "k=1" neighbor value and the "Distance Weight" parameter were selected and the highest success result was obtained.

Implementasi Algoritma Ant Tree Miner Untuk Klasifikasi Jenis Fauna

Classification is a field of data mining that has many methods, one of them is decision tree. Decision tree is proven to be able to classify many kinds of data such as image data and time series data. However, there are several obstacles that are often encountered in the decision tree method. Running time required for the execution of this algorithm is quite long, so this study proposed to use the ant tree miner algorithm which is a development algorithm from the C4.5 decision tree. Ant tree miner works by utilizing ant colony optimization in the process of building its tree structure. Use ant colony optimization expected can optimize the tree that will be formed. From the testing that have been carried out, an accuracy of about 95% is obtained in the process of classifying Zoo dataset with the number of ants between 60 - 90.

Astronomaly: Personalised active anomaly detection in astronomical data

Survey telescopes such as the Vera C. Rubin Observatory and the Square Kilometre Array will discover billions of static and dynamic astronomical sources. Properly mined, these enormous datasets will likely be wellsprings of rare or unknown astrophysical phenomena. The challenge is that the datasets are so large that most data will never be seen by human eyes; currently the most robust instrument we have to detect relevant anomalies. Machine learning is a useful tool for anomaly detection in this regime. However, it struggles to distinguish between interesting anomalies and irrelevant data such as instrumental artefacts or rare astronomical sources that are simply not of interest to a particular scientist. Active learning combines the flexibility and intuition of the human brain with the raw processing power of machine learning. By strategically choosing specific objects for expert labelling, it minimises the amount of data that scientists have to look through while maximising potential scientific return. Here we introduce Astronomaly: a general anomaly detection framework with a novel active learning approach designed to provide personalised recommendations. Astronomaly can operate on most types of astronomical data, including images, light curves and spectra. We use the Galaxy Zoo dataset to demonstrate the effectiveness of Astronomaly, as well as simulated data to thoroughly test our new active learning approach. We find that for both datasets, Astronomaly roughly doubles the number of interesting anomalies found in the first 100 objects viewed by the user. Astronomaly is easily extendable to include new feature extraction techniques, anomaly detection algorithms and even different active learning approaches. The code is publicly available at https://github.com/MichelleLochner/astronomaly.

Spatial clustering analysis of green economy based on knowledge graph

By organically combining principal component analysis, spatial autocorrelation algorithm and two-dimensional graph theory clustering algorithm, the comprehensive evaluation model of regional green economy is explored and established. Based on the evaluation index system of regional green economy, this paper evaluates the development of regional green economy comprehensively by using principal component analysis, and evaluates the competitive advantage of green economy and analyzes the spatial autocorrelation based on the evaluation results. Finally, the green economy and local index score as observed values, by using the method of two-dimensional graph clustering analysis of spatial clustering. In view of the fuzzy k –modes cluster membership degree measure method without considering the defects of the spatial distribution of object, double the distance and density measurement of measure method is introduced into the fuzzy algorithm of k –modes, thus in a more reasonable way to update the membership degree of the object. Vote, MUSH-ROOM and ZOO data sets in UCI machine learning library were used for testing, and the F value of the improved algorithm was better than that of the previous one, indicating that the improved algorithm had good clustering effect. Finally, the improved algorithm is applied to the spatial data collected from Baidu Map to cluster, and a good clustering result is obtained, which shows the feasibility and effectiveness of the algorithm applied to spatial data. Results show that the development of green economy using the analysis method of combining quantitative analysis and qualitative analysis, explores the connotation of green economy with space evaluation model is feasible, small make up for the qualitative analysis of the green economy in the past, can objective system to reflect the regional green economic development level, will help policy makers scientific formulating regional economic development strategy, green integrated development of regional green economy from the macroscopic Angle, the development of network system.

Hexagonality as a New Shape-Based Descriptor of Object

In this paper, we define a new shape-based measure which evaluates how much a given shape is hexagonal. Such an introduced measure ranges through the interval (0, 1] and reaches the maximal possible value 1 if and only if the shape considered is a hexagon. The new measure is also invariant with respect to rotation, translation and scaling transformations. A number of experiments, performed on both synthetic and real image data, are shown in order to confirm theoretical observations and illustrate the behavior of the new measure. The new hexagonality measure also provides several useful side results whose theoretical properties are discussed and experimentally evaluated. As side results, we obtain a new method that computes the shape orientation as the direction which optimizes the new hexagonality measure and a new shape elongation measure which computes the elongation of a given shape as the ratio of the lengths of the longer and shorter semi-axis of the appropriate associated hexagon. Several experiments relating to three well-known image datasets, such as MPEG-7 CE-1, Swedish Leaf, and Galaxy Zoo datasets, are also provided to illustrate effectiveness and benefits of the new introduced shape measures.

Landmark map: An extension of the self-organizing map for a user-intended nonlinear projection

A self-organizing map (SOM) is an unsupervised artificial neural network that is widely used in, e.g., data mining and visualization. Supervised and semi-supervised learning methods have been proposed for the SOM. However, their teacher labels do not describe the relationship between the data and the location of nodes. This study proposes a landmark map (LAMA), which is an extension of SOMs that utilizes several landmarks, e.g., pairs of nodes and data points. LAMA is designed to obtain a user-intended nonlinear projection to achieve, e.g., the landmark-oriented data visualization. To reveal the learning properties of LAMA, the Zoo dataset from the UCI Machine Learning Repository, the McDonald’s dataset from Kaggle, and an artificial formant dataset were analyzed. The analysis results of the Zoo dataset indicated that LAMA could provide a new data view such as the landmark-centered data visualization. McDonald’s dataset analysis demonstrated menu recommendation examples based on a few designated items. Furthermore, the artificial formant data analysis revealed that LAMA successfully provided the intended nonlinear projection associating articular movement with vertical and horizontal movement of a computer cursor. Potential applications of LAMA include data mining, recommendation systems, and human–computer interaction.

Evaluation of H- and G-indices of Scientific Authors using Modified K-Means Clustering Algorithm

In this paper I proposed modified K-means algorithm as the means to assess scientific authors performance by using their h,g-indices values. K-means suffers from poor computational scaling and efficiency as the number of clusters has to be supplied by the user. In this work, I introduce a modification of K-means algorithm that efficiently searches the data to cluster points by compute the sum of squares within each cluster which makes the program to select the most promising subset of classes for clustering. The proposed algorithm was tested on IRIS and ZOO data sets as well as on our local dataset comprising of h- and g-indices, which are the prominent markers for scientific excellence of authors publishing papers in various national and international journals. Results from analyses reveal that the modified k-means algorithm is much faster and outperforms the conventional algorithm in terms of clustering performance, measured by the data discrepancy factor. Index Terms—Modified k-means clustering, h-index, g- index.

Integrated Cluster-based Rule Induction Mining of Temporal Data for Time-Series Analysis

Objectives: To deliver excellent performance when compared to existing method Mining Comprehensible Classification Rules for Time-Series (MCCR-TS) using zoo dataset collected from UCI repository comprising of more than 1000 records. Methods/Analysis: Research works on Time-series data has been evolving as a new trend in current scenarios due to the wide range of applications involved in it. One of the widely researched topics is the web usage mining and effective evaluation of time-series movements and transaction associated with them. Most of the studies conducted earlier, focused on identifying the time-series data from the entire logs. However, this type of patterns may not be accurate enough for evaluation due to differentiated behaviors of user patterns are not taken into consideration. Findings: In this paper, we examine the time series for temporal data using integrated cluster-based rule induction mining, by presenting a novel algorithm, namely, Integrated Cluster-based Rule Induction Mining for Time-Series analysis (ICRIM-TS). The algorithm discovers temporal data and the similarities between users are evaluated by the proposed measure, Time Series. To our best knowledge, this is the first work on mining and identification of user behaviors based on the time series data with preferences being given for both user relations and temporal data at the same time. Improvement: Through experimental evaluation, under various settings, the performance of web usage mining is evaluated in terms of precision of users, error threshold value and number of rules extracted.

Evaluation of h- and g-indices of Scientific Authors using Modified K-Means Clustering Algorithm

<p>In this paper we proposed modified K-means algorithm to assess scientific authors performance by using their h,g-indices values. K-means suffers from poor computational scaling and efficiency as the number of clusters has to be supplied by the user. Hence, in this work, we introduce a modification of K-means algorithm that efficiently searches the data to cluster points by compute the sum of squares within each cluster which makes the program to select the most promising subset of classes for clustering. The proposed algorithm was tested on IRIS and ZOO data sets as well as on our local dataset comprising of h- and g-indices, which are the prominent markers for scientific excellence of authors publishing papers in various national and international journals. Results from analysis reveal that the modified k-means algorithm is much faster and outperforms the conventional algorithm in terms of clustering performance, measured by the data discrepancy factor.</p>

SVM-RFE based feature selection and Taguchi parameters optimization for multiclass SVM classifier.

Recently, support vector machine (SVM) has excellent performance on classification and prediction and is widely used on disease diagnosis or medical assistance. However, SVM only functions well on two-group classification problems. This study combines feature selection and SVM recursive feature elimination (SVM-RFE) to investigate the classification accuracy of multiclass problems for Dermatology and Zoo databases. Dermatology dataset contains 33 feature variables, 1 class variable, and 366 testing instances; and the Zoo dataset contains 16 feature variables, 1 class variable, and 101 testing instances. The feature variables in the two datasets were sorted in descending order by explanatory power, and different feature sets were selected by SVM-RFE to explore classification accuracy. Meanwhile, Taguchi method was jointly combined with SVM classifier in order to optimize parameters C and γ to increase classification accuracy for multiclass classification. The experimental results show that the classification accuracy can be more than 95% after SVM-RFE feature selection and Taguchi parameter optimization for Dermatology and Zoo databases.

Fast progressive training of mixture models for model selection

Finite mixture models (FMM) are flexible models with varying uses such as density estimation, clustering, classification, modeling heterogeneity, model averaging, and handling missing data. Expectation maximization (EM) algorithm can learn the maximum likelihood estimates for the model parameters. One of the prerequisites for using the EM algorithm is the a priori knowledge of the number of mixture components in the mixture model. However, the number of mixing components is often unknown. Therefore, determining the number of mixture components has been a central problem in mixture modelling. Thus, mixture modelling is often a two-stage process of determining the number of mixture components and then estimating the parameters of the mixture model. This paper proposes a fast training of a series of mixture models using progressive merging of mixture components to facilitate model selection algorithm to make appropriate choice of the model. The paper also proposes a data driven, fast approximation of the Kullback---Leibler (KL) divergence as a criterion to measure the similarity of the mixture components. We use the proposed methodology in mixture modelling of a synthetic dataset, a publicly available zoo dataset, and two chromosomal aberration datasets showing that model selection is efficient and effective.

러프 엔트로피를 이용한 범주형 데이터의 클러스터링

A variety of cluster analysis techniques prerequisite to cluster objects having similar characteristics in data mining. But the clustering of those algorithms have lots of difficulties in dealing with categorical data within the databases. The imprecise handling of uncertainty within categorical data in the clustering process stems from the only algebraic logic of rough set, resulting in the degradation of stability and effectiveness. This paper proposes a information-theoretic rough entropy(RE) by taking into account the dependency of attributes and proposes a technique called min-mean-mean roughness(MMMR) for selecting clustering attribute. We analyze and compare the performance of the proposed technique with K-means, fuzzy techniques and other standard deviation roughness methods based on ZOO dataset. The results verify the better performance of the proposed approach.