Knowledge Discovery Algorithms Research Articles

Toward a campus crisis management system amid the pandemic and beyond

This study develops artifact and design principles for a campus-wide crisis management system amid the COVID-19 pandemic and beyond. Drawing upon crisis management literature and the uncertainty reduction theory, we develop a preliminary crisis management system with three fundamental components: dashboard, portal, and forum, geared toward aggregating environmental information, expert knowledge, and stakeholder insights. This immediate, integrated, and interactive campus crisis management system aims to promote organizational learning through knowledge acquisition, dissemination, and utilization while reducing stakeholders’ uncertainty about the crisis. Through comprehensive evaluation, we further synthesize and summarize new design principles for prospective artifacts, including constructing a distributed crisis management network, capitalizing on data mining and knowledge discovery algorithms, and developing the adaptability of the crisis management system in various circumstances.

Unsupervised learning: Local and global structure preservation in industrial data

This article examines the use of data mining and machine learning algorithms for knowledge discovery in industrial processes, with a focus on understanding how these methods distinguish and represent the global and local structures underlying data. The effectiveness of alternative dimensionality reduction algorithms in characterizing process operations is also explored, with an emphasis on a novel dimensionality reduction method, PaCMAP, compared to the dependable historical method of PCA and the more recent t-SNE. The article demonstrates a visualization method for identifying the variables responsible for the separation between selected clusters and studies the effects of key elements of the algorithms on cluster analysis in light of domain knowledge. Results from an industrial case study of a pyrolysis reactor validates the applicability of these methods in real-life scenarios. Overall, this article provides insights into the use of machine learning and dimensionality reduction algorithms for knowledge discovery in industrial processes.

Automatic clustering and feature selection using multi-objective crow search algorithm

Today’s real-world data is frequently significant in size, with many redundant, missing, and noise-based features and data instances must be addressed before applying various data-mining-based algorithms for further knowledge discovery. Excessive dimensionality may be mitigated by carefully excluding unnecessary characteristics and selecting a reasonable subset of features. When presented as an optimization issue, choosing the best clusters using the most suitable subset of attributes is a challenge that may be handled using practical meta-heuristic approaches. Besides this, the automatic finding of the appropriate cluster number is another challenging task for the real-world dataset in the unsupervised machine-learning study. The present work proposes a multi-objective crow search algorithm for clustering and feature selection (MO-CSACFS) by modifying the crow search algorithm and introducing a levy flight-based two-point cross-over mechanism for a better exploration phase of the crow and further making it suitable for multi-objective optimization problems. MO-CSACFS addresses both issues using the three objective functions to find appropriate cluster numbers and features. MO-CSACFS is implemented over several real-life and synthetic datasets with varying instances, features, and cluster numbers to assess the algorithm’s performance; apart from that, the present work is also applied over several gene-expression datasets. MO-CSACFS is compared with two similar recently proposed multi-objective optimization processes used over an automatic, unsupervised machine learning task. The results show that the MO-CSACFS has produced a compact and robust cluster comparable to other similar works from the literature.

CAC-WOA: context aware clustering with whale optimization algorithm for knowledge discovery from multidimensional space in electricity application

Energy consumption forecasting is a hot field of research; despite the number of developed models, projecting electric consumption in residential buildings remains problematic owing to the significant unpredictability of occupant energy use behavior. Discovering the electricity consumption knowledge from the multi-dimensional data streams (MDDS) of electricity logs is a challenging research problem. We propose a novel electricity knowledge discovery model proposed from the MDDS using clustering and machine learning. Context-aware clustering with whale optimization algorithm (CAC-WOA) is proposed to discover the predictive features from the electricity MDDS and perform the predictions using WOA. The CAC-WOA consists of two phases context-aware group formation and a WOA-based machine learning predictive model. In the CAC algorithm, group formation using electricity contextual information to estimate the robust predictive features are proposed. Using such predictive features, the predictive model using the WOA-based artificial neural network (ANN) is built. The modified ANN technique using the WOA algorithm is used to reduce the error rates and improve the prediction accuracy. The experimental outcomes using publicly available electricity consumption datasets prove the efficiency of the CAC-WOA model. Overall prediction accuracy is improved by 3.27% and prediction time is reduced by 11.31% using CAC-WOA compared state-of-the-art solutions.

Methods and Algorithms for Data Analysis and Knowledge Discovery from Data

Methods and Algorithms for Data Analysis and Knowledge Discovery from Data

Balancing between holistic and cumulative sentiment classification

Sentiment analysis is a fast-accelerating discipline that develops algorithms for knowledge discovery from opinionated content. The challenges however, when it comes to analyzing user reviews are plenty. Bad-quality, informal use of language and lack of labels, are only a few obstacles. Most importantly, users, consciously or subconsciously, use different approaches for expressing their opinion about a product or a service. Some of them go sentence by sentence mentioning some positive and negative aspects whereas others provide a mixed piece of text where the reader is supposed to see the big picture to understand the message. In this work, we propose a novel neural network that deals with both situations. Our method, by combining convolutional, recurrent and attention neural networks can extract rich linguistic patterns that reveal the user’s sentiment towards the entity under review. We evaluate our method in nine datasets that represent both binary and multi-class classification tasks. Experimental evaluation indicates that our method outperforms well-established deep learning approaches. Our approach outperformed the competitive methods in 8 out of 9 cases.

An Evolutive Frequent Pattern Tree-based Incremental Knowledge Discovery Algorithm

To understand current situation in specific scenarios, valuable knowledge should be mined from both historical data and emerging new data. However, most existing algorithms take the historical data and the emerging data as a whole and periodically repeat to analyze all of them, which results in heavy computation overhead. It is also challenging to accurately discover new knowledge in time, because the emerging data are usually small compared to the historical data. To address these challenges, we propose a novel knowledge discovery algorithm based on double evolving frequent pattern trees that can trace the dynamically evolving data by an incremental sliding window. One tree is used to record frequent patterns from the historical data, and the other one records incremental frequent items. The structures of the double frequent pattern trees and their relationships are updated periodically according to the emerging data and a sliding window. New frequent patterns are mined from the incremental data and new knowledge can be obtained from pattern changes. Evaluations show that this algorithm can discover new knowledge from evolving data with good performance and high accuracy.

Knowledge discovery using an enhanced latent Dirichlet allocation-based clustering method for solving on-site assembly problems

• We propose a text mining approach combining topic modeling, text clustering , and association rule identification algorithms for knowledge discovery. • We propose an enhanced latent Dirichlet allocation scheme to explore document-topic and topic-word distributions. • We introduce a refined DBSCAN method for text clustering. • The results demonstrate the capability and effectiveness of the proposed approach. Prompt responses to problems/faults arising in an assembly workshop are crucial in terms of production reliability and efficiency. However, human-dependent tasks are time-consuming and prone to error. In this paper, we propose a knowledge discovery approach. We extract the patterns of associations between texts in problem-solving records to generate appropriate solutions automatically. First, we use an enhanced latent Dirichlet allocation (EnLDA) technique to explore the document-topic and topic-word distributions of a text corpus recording assembly problems, causes, and solutions. To increase accuracy, we adjust the elements of the document-term matrix, and we assign term frequency-inverse document frequencies. Second, we use the Refining Density-based Spatial Clustering of Application with Noise (Rf-DBSCAN) algorithm for text clustering. This refines the distances among topic distribution vectors and incorporates noise objects into clustering. This clusters textual documents with similar semantic information, maximizing information retention. Third, we use the Apriori algorithm to identify pattern associations among document clusters that represent the problems, causes, and solutions. We perform a case study using field data from an automobile assembly workshop. The results show that the method retrieves hidden but valuable information from textual records. The decision support knowledge facilitates assembly problem-solving.

Surrogate-assisted performance prediction for data-driven knowledge discovery algorithms: Application to evolutionary modeling of clinical pathways

The paper proposes and investigates an approach for surrogate-assisted performance prediction of data-driven knowledge discovery algorithms. The approach is based on the identification of surrogate models for prediction of the target algorithm’s quality and performance. The proposed approach was implemented and investigated as applied to an evolutionary algorithm for discovering clusters of interpretable clinical pathways in electronic health records of patients with acute coronary syndrome. Several clustering metrics and execution time were used as the target quality and performance metrics respectively. An analytical software prototype based on the proposed approach for the prediction of algorithm characteristics and feature analysis was developed to provide a more interpretable prediction of the target algorithm’s performance and quality that can be further used for parameter tuning.

An Improved Evolutionary Algorithm for Data Mining and Knowledge Discovery

Recent advancements in computer technologies for data processing, collection, and storage have offered several chances to improve the abilities in production, services, communication, and researches. Data mining (DM) is an interdisciplinary field commonly used to extract useful patterns from the data. At the same time, educational data mining (EDM) is a kind of DM concept, which finds use in educational sector. Recently, artificial intelligence (AI) techniques can be used for mining a large amount of data. At the same time, in DM, the feature selection process becomes necessary to generate subset of features and can be solved by the use of metaheuristic optimization algorithms. With this motivation, this paper presents an improved evolutionary algorithm based feature subsets election with neuro-fuzzy classification (IEAFSS-NFC) for data mining in the education sector. The presented IEAFSS-NFC model involves data pre-processing, feature selection, and classification. Besides, the Chaotic Whale Optimization Algorithm (CWOA) is used for the selection of the highly related feature subsets to accomplish improved classification results. Then, Neuro-Fuzzy Classification (NFC) technique is employed for the classification of education data. The IEAFSS-NFC model is tested against a benchmark Student Performance DataSet from the UCI repository. The simulation outcome has shown that the IEAFSS-NFC model is superior to other methods.

Detection and interpretation of erroneous data in statistical analysis of consumption of energy resources

Monitoring and analysis of consumption of energy resources in various contexts, as well as measuring of parameters (indicators) in time are of utmost importance for the modern economy. This work is dedicated to examination and interpretation of the anomalies of collecting data on consumption of energy resources (on the example of gas consumption) in the municipal formation. Gas consumption is important for the socioeconomic sphere of cities. Unauthorized connections are the key reason for non-technological waste of the resource. The traditional methods of detection of stealing of gas are ineffective and time-consuming. The modern technologies of data analysis would allow detecting and interpreting the anomalies of consumption, as well as forming the lists for checking the objects for unauthorized connections. The author’s special contribution lies in application of the set of statistical methods aimed at processing and identification of anomalies in energy consumption of a municipal formation. It is worth noting that the use of such technologies requires the development of effective algorithms and implementation of automation and machine learning algorithms. The new perspective upon time-series data facilitates identification of anomalies, optimization of decision-making, etc. These processes can be automated. The presented methodology tested on time-series data that describes the consumption of gas can be used for a broader range of tasks. The research can be combined with the methods of knowledge discovery and deep learning algorithms.

Extracting and inserting knowledge into stacked denoising auto-encoders

Deep neural networks (DNNs) with a complex structure and multiple nonlinear processing units have achieved great successes for feature learning in image and visualization analysis. Due to interpretability of the “black box” problem in DNNs, however, there are still many obstacles to applications of DNNs in various real-world cases. This paper proposes a new DNN model, knowledge-based deep stacked denoising auto-encoders (KBSDAE), which inserts the knowledge (i.e., confidence and classification rules) into the deep network structure. This model not only can offer a good understanding of the representations learned by the deep network but also can produce an improvement in the learning performance of stacked denoising auto-encoder (SDAE). The knowledge discovery algorithm is proposed to extract confidence rules to interpret the layerwise network (i.e., denoising auto-encoder (DAE)). The symbolic language is developed to describe the deep network and shows that it is suitable for the representation of quantitative reasoning in a deep network. The confidence rule insertion to the deep network is able to produce an improvement in feature learning of DAEs. The classification rules extracted from the data offer a novel method for knowledge insertion to the classification layer of SDAE. The testing results of KBSDAE on various benchmark data indicate that the proposed method not only effectively extracts knowledge from the deep network, but also shows better feature learning performance than that of those typical DNNs (e.g., SDAE).

A Data-Driven Method for Power System Transient Instability Mode Identification Based on Knowledge Discovery and XGBoost Algorithm

Aiming at the difficulty of unstable pattern recognition after power system fault, a novel identification framework for transient instability mode identification based on knowledge discovery by accuracy maximization (KODAMA) and extreme gradient boosting (XGBoost) algorithm is proposed. In this method, the transient stability of all typical fault scenarios of power system is obtained firstly by XGBoost. Then, to make full use of the structure of the raw data and mine the contained data information, a novel data mining algorithm KODAMA is introduced to cluster the mode of rotor angle in case of instability, thus to convert pattern-unlabeled case data into pattern-labeled data. Finally, based on this labeled data, to fully reflect the dynamic characteristics, a multiple XGBoost assessment strategy is designed to recognize different instable modes. The proposed technique is tested on the Nordic test system, and the results indicate that the proposed approach can provide fast and accurate recognition of instable mode and has a certain prospect of online application.

Artificial Intelligence and Special Rraining of Higher Education Talents in Response to Public Health Emergencies from the Perspective of Psychology

The development of artificial intelligence technology brings new opportunities for social governance, especially in the emergency treatment of public health emergencies; the application scenarios of artificial intelligence are more and more abundant. Based on the study of public health emergencies from the perspective of psychology, this paper systematically studies the cultivation of talents in higher education and the application of artificial intelligence by using qualitative and quantitative research methods such as literature analysis and questionnaire survey. In this paper, we use Chinese and foreign academic journal network, China master and doctoral dissertation database, VIP and other network resource databases to search for literature and materials related to the quality of higher education and the application of artificial intelligence, and summarize and sort out the relevant data, so as to find out the shortcomings of existing research. The results show that "innovation ability" is only 3.29, "practice and skills" accounts for 38%. Colleges and universities should strengthen the cultivation of students' innovative ability and medical workers' practice and skills, so as to better respond to public health emergencies. The experimental results of drug solubility prediction show that the prediction efficiency of neural network is better than that of linear regression model. Big data driven artificial intelligence knowledge discovery algorithm is becoming an effective tool to improve drug development efficiency and reduce research and development costs.

Orbit Computation for Atomically Generated Subgroups of Isometries of $\mathbb{Z}^n$

Partitions of a set play an important role in automatic knowledge discovery algorithms that learn human-like concepts from human-like innate priors. One such type of prior is isometries, the recogn...

A Hybrid Data Reduction and Knowledge Extraction Algorithm for Quality Prediction

Background: With the explosive growth of the manufacturing data, the manufacturing enterprises paid more and more attention to dealing with the manufacturing big data. The manufacturing big data also can be summarized as "5Vs”, volume, variety, velocity, veracity and value. Recently, the researchers are focused on proposing better knowledge discovery algorithms to handling the manufacturing big data. Objective: The high dimensional data can be reduced from two directions. The one was the dimension reduction. It makes the data set simple and overcome the problem of curse dimensionality. This method reduced the data set form the data width. Methods: We proposed a hybrid data reduction and knowledge extraction algorithm (HDRKE) for quality prediction. There are 5 steps in the algorithm: Step 1: Data preprocessing; Step 2: Dimension reduction; Step 3: Extract SVs by SVM; Step 4: Extract rules from the subset; Step 5: Prediction by the rules extracted in step 3. Results: The presented HDRKE method reduced the data scales from the data dimensions and the data attributions. Then, the prediction method was used on the subset of reduced data. At last, the HDRKE method was applied to a enterprise sample, the validation of the method can be validated on the enterprise sample. Conclusion: Quality prediction and control was an important procedure in manufacturing. The HDRKE algorithm was a novel method based on the attribution reduction and dimensionality reduce. The data set simplified from double direction made the data set easily to calculate. The HDRKE method also proposed a new thought of decision rules extracting on the low-embeddings. The HDRKE method also applied to a manufacturing instance and proved its validity.

Designing of fog based FBCMI2E Model using machine learning approaches for intelligent communication systems

The work discusses the evolution of communication models and technological aspects for developing inclusive platforms. The paper gives illustrations and graphical presentations about components of the inclusive platform. An inclusive platform consists of fog devices that can collect the fitness data, geospatial data and mobile call details from the people who are in need of help from the government agencies. The paper discusses the controversies around the definition and qualification of the informal economy as well as presents a simulated scenario. The case presented here gives an insight on how fog devices and data mining can be used for bringing people into main fold of the economy. The inclusiveness index of the person is computed on the basis of four aspects. The first aspect is the fitness, the second is his inner social circle, third is her/his reliability to remain in a place, and last is call analysis. While computing the fitness index, it was found that Naive Bayes (NB) algorithm has the maximum accuracy with respect to K-Nearest Neighbors (KNN), Decision Tree (DT) and Linear Discriminant Analysis (LDA). For computing inner social circle, Louvain algorithm helped to compute stability and strength of socio-economic ties of the individual. For geospatial and call analysis, insights from knowledge discovery algorithm such as FP-Growth helped to arrive at decision to qualify the person for inclusive program. The paper ends with details on how to automate the inclusiveness index computation using neural network. The research indicates that energy is the key constraint for implementing such programs. Hence, a theoretical analysis about energy efficiency is also explained in the paper.

A Metropolis within Gibbs algorithm for knowledge discovery in language assessments

The reduced reparameterised unified model (RRUM) has been frequently used in language assessment studies. The objective of this research is to advance an MCMC algorithm for the Bayesian RRUM. The algorithm starts with estimating correlated attributes. Using a saturated model and a binary decimal conversion, the algorithm transforms possible attribute patterns to a multinomial distribution. Along with the likelihood of an attribute pattern, a Dirichlet distribution is used as the prior to sample from the posterior. The Dirichlet distribution is constructed using Gamma distributions. Correlated attributes of examinees are generated using the inverse transform sampling. Model parameters are estimated using the Metropolis within Gibbs sampler sequentially. Two simulation studies are conducted to evaluate the performance of the algorithm. The simulation studies show that as sample size increases, the measure of accuracy increases. The algorithm developed in this research is implemented in R.

A Metropolis within Gibbs algorithm for knowledge discovery in language assessments

The reduced reparameterised unified model (RRUM) has been frequently used in language assessment studies. The objective of this research is to advance an MCMC algorithm for the Bayesian RRUM. The algorithm starts with estimating correlated attributes. Using a saturated model and a binary decimal conversion, the algorithm transforms possible attribute patterns to a multinomial distribution. Along with the likelihood of an attribute pattern, a Dirichlet distribution is used as the prior to sample from the posterior. The Dirichlet distribution is constructed using Gamma distributions. Correlated attributes of examinees are generated using the inverse transform sampling. Model parameters are estimated using the Metropolis within Gibbs sampler sequentially. Two simulation studies are conducted to evaluate the performance of the algorithm. The simulation studies show that as sample size increases, the measure of accuracy increases. The algorithm developed in this research is implemented in R.

Finding strongly connected components of simple digraphs based on granulation strategy

Strongly connected components (SCCs) are an important kind of subgraphs in digraphs. It can be viewed as a kind of knowledge in the viewpoint of knowledge discovery. In our previous work, a knowledge discovery algorithm called RSCC was proposed for finding SCCs of simple digraphs based on two operators, k-step R-related set and k-step upper approximation, of rough set theory (RST). RSCC algorithm can find SCCs more efficiently than Tarjan algorithm of linear complexity. However, on the one hand, as the theoretical basis of RST applied to SCCs discovery of digraphs, the theoretical relationships between RST and graph theory investigated in previous work only include four equivalences between fundamental RST and graph concepts relating to SCCs. The reasonability of using the two RST operators to find SCCs still need to be investigated. On the other hand, it is found that there are three SCCs correlations between vertices after we use three RST concepts, R-related set, lower and upper approximation sets, to analyze SCCs. RSCC algorithm ignores these SCCs correlations so that the efficiency of RSCC is affected negatively. For the above two issues, firstly, we explore the equivalence between the two RST operators and Breadth-First Search (BFS) which is one of the most basic graph search algorithms and the most direct way to find SCCs. These equivalences explain the reasonability of using the two RST operators to find SCCs, and enrich the content of the theoretical relationships between RST and graph theory. Secondly, we design a granulation strategy according to these three SCCs correlations. Then an algorithm called GRSCC for finding SCCs of simple digraphs based on granulation strategy is proposed. Experimental results show that GRSCC provides better performance to RSCC.