Research on Network Data Algorithm Based on Association Rules

Aiming at the key problems of poor real-time correlation mining of massive operation data of multimedia software-defined network (SDN) and low accuracy of situation prediction, an intelligent mining method of multimedia software-defined network alarm knowledge is proposed to ensure that the network fault diagnosis rule base can evolve and improve continuously based on network changes, so as to improve the effectiveness and reliability of network fault diagnosis. Specifically, this method first uses the algorithm of generating second-order frequent itemset association rules based on the matrix to generate fault alarm association rules, and then deeply analyzes the mining results based on the existing models in the rule base, including conflict detection, conflict processing, and multiset rule derivation, Conflict detection is responsible for detecting whether there is a conflict between the association rules generated by the new mining and the existing association rules in the knowledge base. Conflict handling is responsible for orderly importing or discarding the detected conflict rules. Multiset rule derivation is responsible for obtaining the relationship of deeper data items that multiple association rules depend on each other. Finally, the implementation of the system proves the application value of this method

Association rule mining among itemsets is a fundamental task and is of great importance in many data mining applications including attacks in network data, stock market, financial applications, bioinformatics to find genetic disorders, etc. However, association rule extraction from a reasonable-sized database produces a large number of rules. As a result, many of them are redundant to other rules, and they are practically useless. To overcome this issue, methods for mining non-redundant rules are essentially required. To address such problem, we initially propose a definition for redundancy in sense of minimal knowledge and then a compact representation of non-redundant association rules which we call as compact informative generic basis. We also provide an improved version of the existing DCI_CLOSED algorithm (DCI_PLUS) to find out the frequent closed itemsets (FCI) with their minimal representative generators in combination with BitTable which represents a compact database form in a single scan of the original database. We further introduce an algorithm for constructing the compact informative generic basis from the FCI and their generators in an efficient way. We finally present an inference mechanism in which all association rules can be generated without accessing the database. Experiments are performed on the proposed method. The experimental results show that the proposed method outperforms the other existing related methods.

Data mining involves the use of advanced data analysis tools to find out new, suitable patterns and project the relationship among the patterns which were not known prior. In data mining, association rule learning is a popular and well researched method for discovering interesting relations between variables in large databases. One of the emerging research areas under Data mining is Social Networks. The objective of this paper focuses on the formulation of association rules using which decisions can be made for future Endeavour. This research applies Apriori Algorithm which is one of the classical algorithms for deriving association rules. The Algorithm is applied to Face book 100 university dataset which has originated from Adam D’Angelo of Face book. It contains self-defined characteristics of a person including variables like residence, year, and major, second major, gender, school. This paper to begin with the research uses only ten Universities and highlights the formation of association rules between the attributes or variables and explores the association rule between a course and gender, and discovers the influence of gender in studying a course. The previous research with this dataset has applied only regression models and this is the first time to apply association rules.

It is difficult to mine association rules between multi-source heterogeneous network nodes, which leads to the high false detection rate, node location error, and location overhead of multi-source heterogeneous network abnormal nodes. Therefore, this article proposes a new abnormal node location algorithm in multi-source heterogeneous network based on association rules. The algorithm of sliding windows multiple value pattern tree is used to extract the association of multi-source heterogeneous network data and realize the mining of association rules between multi-source heterogeneous network nodes. The association rule mining results are preprocessed, and the multi-source heterogeneous network abnormal node location model is built using the multidimensional scaling algorithm and the low rank matrix decomposition model. The model is solved to obtain the multi-source heterogeneous network abnormal node location results. The experimental results show that the false detection rate of abnormal nodes in the proposed algorithm is always below 4.6 %, the maximum node location error is only 2.1 mm, and the average location overhead is 55.1 ms, which can achieve the goal of fast and accurate location of abnormal nodes in multi-source heterogeneous networks.

Association rule mining has been a highly active research field over the past decade. Extraction of frequency-related patterns has been applied to several domains. However, the way association rules are defined has limited people’s ability to obtain all the patterns of interest. In this chapter, the authors present an alternative approach that allows us to obtain new kinds of association rules that represent deviations from common behaviors. These new rules are called anomalous rules. To obtain such rules requires that we extract all the most frequent patterns together with certain extension patterns that may occur very infrequently. An approach that relies on anomalous rules has possible application in the areas of counterterrorism, fraud detection, pharmaceutical data analysis and network intrusion detection. They provide an adaption of measures of interest to our anomalous rule sets, and we propose an algorithm that can extract anomalous rules as well. Their experiments with benchmark and real-life datasets suggest that the set of anomalous rules is smaller than the set of association rules. Their work also provides evidence that our proposed approach can discover hidden patterns with good reliability.

We investigate the problem of tuning and selecting among interestingness measures for association rules. We first derive a parametric normalization factor for such measures that addresses imbalanced itemset sizes, and show how it can be generalized across many previously derived measures. Next, we develop a validationbased framework for both the normalization and selection tasks, based upon mutual information measures over attributes. We then apply this framework to market basket data and user profile data in weblogs, to automatically choose among or fine-tune alternative measures for generating and ranking rules. Finally, we show how the derived normalization factor can significantly improve the sensitivity of interestingness measures when used for pure association rule mining and also for a classification task. We also consider how this data-driven approach can be used for fusion of association rule sets: either those elicited from subject matter experts, or those found using prior background knowledge. INTRODUCTION One of the most important aspects of association rule mining is ranking rules by their significance, according to some quantitative measure that expresses their interestingness with respect to a decision support or associative reasoning task. Rules take the form X → Y, where both X and Y are subsets of an observed itemset L = {I1, I2, ..., Ik}. Two well-known measures for association rule interestingness are the support, P(X) and the confidence, P(Y | X). These probabilistic measures have been used with other statistical formulae to derive compound measures used in discovering the most significant rule. One limitation of existing binary measures of rule interestingness is that they do not account for the relative size of the itemsets to which each candidate pair of associated subsets (X, Y) belongs. Moreover, there are some hidden associations related to candidates appearing in small groups. Thus, giving some attention and weight to these small groups may lead us to a different relationship perspective. This kind of data behavior can be seen, for example, in social network data where each user record consists of features such as interests, communities, schools attended, etc. In particular, user’s list of interests, each of which corresponds to a list of interest holders. Some interests such as “DNA replication” have low membership; whether this is because the interests are less popular or more specialized, it often suggests a more significant association between users naming them than between those who have interests such as “Music” or “Games” in

In view of the problems of poor dynamic clustering effect and long time consuming in traditional clustering methods, a new method of dynamic clustering of network data based on association rules is proposed. Through the analysis of network data attributes, the association rules are established, and based on this, the value function of network data is determined. Finally, dynamic clustering of network data is realized when the clustering constraints are met. The experimental results show that the improved method can effectively improve the purity of clustering, and has some advantages in improving the clustering effect.

Under the guidance of analogical thinking in traditional Chinese medicine (TCM), the "treating of skin with skin" therapy, using processed animal and plant skin-derived medicinal materials to treat skin diseases, has a long history but lacks scientific evidence-based support. This study aims to apply data mining and network pharmacology techniques to explore the prescription patterns and mechanisms of action of the "treating skin with skin" therapy, and to interpret its rationality, effectiveness, and scientific basis using modern scientific methods. Relevant literature from Chinese and English databases over the past 20 years was retrieved and integrated according to inclusion and exclusion criteria. Data were integrated and mined using software such as Excel, OriginPro 2021, and SPSS Modeler 18.0. Frequency analysis, cluster analysis, and association rule analysis were performed sequentially to summarize high-frequency plant skin-derived Chinese medicinal materials and extract the core prescription. Core prescription drugs and skin disease-related gene targets were obtained from drug and disease target databases, including the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform and GeneCards. Intersection targets were identified using Venny 2.1.0, and core targets were further extracted. Protein-protein interaction (PPI) network analysis, Gene Ontology (GO) functional enrichment analysis, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were conducted for the intersection targets, and finally a "core drug-component-target-pathway" network was constructed and visualized. Based on the inclusion and exclusion criteria, 563 eligible articles were ultimately included, from which 32 prescriptions were extracted, involving 21 plant skin-derived Chinese medicinal materials. Among them, 15 medicinal materials had a usage frequency exceeding 50%, with bitter, sweet, and cold properties predominating. Cluster analysis yielded 4 effective clusters. Cluster 1: Phellodendron cortex-Dictamni cortex; Cluster 2: Eucommia cortex-Mori cortex-Meliae cortex-Pseudolaricis cortex; Cluster 3: Moutan cortex-Poriae Cutis-Lycii cortex-Fraxini cortex-Acanthopanacis cortex-Benincasae Pericarpium; Cluster 4: Ailanthi cortex-Periplocae cortex-Erycibes cortex. Association rule analysis generated 92 association rules, with 4 strong links, 90 moderate links, and 10 weak links between herb pairs. Based on the association rules and linkage strength, the core prescription of the "treating skin with skin" therapy was summarized as "Phellodendri Cortex-Moutan Cortex-Dictamni Cortex-Benincasae Pericarpium-Poriae Cutis-Lycii Cortex." A total of 46 active ingredients were screened from this core prescription, involving 781 gene targets. A total of 2 537 skin disease-related gene targets were retrieved, yielding 212 intersection targets between the core prescription and the disease. PPI network analysis of the intersection targets showed 2 723 edges among 212 nodes, with an average node degree of 25.9 and an average local clustering coefficient of 0.554. The PPI network of core targets contained 50 nodes and 784 edges; the top three core nodes by degrees were interleukin-6 (IL-6; 116), tumor necrosis factor (TNF; 115), and interleukin-1 beta (IL-1β; 110). GO analysis yielded 732 biological process terms, with the top 3 being response to lipopolysaccharide, positive regulation of DNA-templated transcription, and positive regulation of microRNA (miRNA) transcription; 90 cellular component terms, with the top 3 being extracellular space, extracellular region, and chromatin; and 153 molecular function terms, with the top 3 being enzyme binding, nuclear receptor activity, and identical protein binding. KEGG analysis identified 166 enriched signaling pathways, with the top 3 being pathways in cancer, lipid and atherosclerosis, and the advanced glycation end products-receptor for advanced glycation end products pathway in diabetic complications. The visualized "core drug-component-target-pathway" network comprised 854 nodes and 1 911 edges, with a network centralization coefficient of 0.182 and a characteristic path length of 3.985; the top 3 nodes by degree were all drug components, namely vitamin E, vitamin B, and trigonelline. Guided by TCM analogical thinking, the core prescription of the "treating skin with skin" therapy exerts therapeutic effects on skin diseases through synergistic actions involving multiple components, multiple targets, and multiple pathways. Its effect is significant, the rationale is well founded, and its origins are well established.

In a local area network (LAN) where users are relatively stable, the usage patterns of systems and working habits are also stable. This character implies that there exist many rules in corresponding network applications. By intercepting all the frames in the LAN and pre-processing the collected data, association rules mining techniques can be used to extract association rules from the network data. These rules can latter be effectively applied to network anomaly detection in the LAN. This paper discusses the method of using association rules mining in anomaly detection of LAN, and analyzes its working principle. We give a detailed discussion on several steps, including the method of data acquisition and preprocessing, association rules mining, the usage of similarity to determine whether the network behavior conform to the extracted association rules and the detection of anomalous behaviors. Finally, the corresponding experimental results are given.

In the intelligent distribution network, various faults of the distribution network still frequently occur during the operation of the distribution network, which normally appears at the weak points of various types of electrical equipment. The basic concepts and practical application significance of the three judgment indicators of support, confidence, and imbalance ratio in association rules are proposed in this paper; subsequently, the collection and sorting of fault data in the distribution network are mentioned, and the subsequent fault data of the association rules are performed as well. The preprocessing was implemented, and the weak point analysis model of the distribution network was established. Finally, the analysis and verification were carried out through the calculation example, which proved the feasibility and accuracy of the association rule data mining algorithm FP-network in the analysis and diagnosis of the weak point of the distribution network. The results show that the accuracy of the proposed method in identifying weak points in the distribution network can reach more than 90%. Therefore, the application of this algorithm in the monitoring of weak points in the distribution network has better performance.

Association rules discovered through attribute-oriented induction are commonly used in data mining tools to express relationships between variables. However, causal inference algorithms discover more concise relationships between variables, namely, relations of direct cause. These algorithms produce regressive structured equation models for continuous linear data and Bayes networks for discrete data. This work compares the effectiveness of causal inference algorithms with association rule induction for discovering patterns in discrete data.KeywordsBayesian NetworkAssociation RuleMarkov ConditionContraceptive UsageAssociation Rule GenerationThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Replication is one of the mechanisms managing data since it improves data access and reliability. However, in recent years, with widely available, low-cost technology, the amount of various data grows rapidly. The problem is although we are packed with data, but we have lacked of knowledge. Nevertheless, if the impractical data is used in database replication, this will cause waste of data storage and the time taken for a replication process will be delayed. This paper propose a new algorithm namely Binary Vote Assignment on Grid Quorum with Association Rule (BVAGQ-AR) in order to handle fragmented database replication. BVAGQ-AR algorithm is capable of partitioning the database into disjoint fragments. Fragmentation in distributed database is very useful in terms of usage, reliability and efficiency. Handling fragmented database replication becomes challenging issue to administrator since the distributed database is scattered into split replica partitions or fragments. We address how to build reliable system by using the proposed BVAGQ-AR algorithm for distributed database fragmentation by using Malaysian Greater Research Network (MyGRANTS) data as a case study. The result shows that managing fragmented database replication through proposed BVAGQ-AR algorithm able to preserve MyGRANTS data consistency.

With the development of network technology, decision support and data mining has been widely applied in many fields, such as finance, marketing, quality analysis. But it isn't applied widely in the evaluation of network education. In order to improve the quality of personal training, universities regard the “the quality of teaching and teaching reform project” as the starting point, carry out teaching and management reform and explore the new talent cultivation mode. This paper summarizes the relative theories and methods of decision support, then introduces association rule in data mining. First, we analyze the classical association rules mining Apriori algorithm, then analyse this algorithm's disadvantages and improvement ways. Second, we use the Apriori algorithm, which can evaluate the network education from different aspects, thus providing manager with clear and actionable feedback to enhance their practice. Finally, the algorithm's efficiency is assessed and we points out the further research problems.

Though widely used in the treatment of coronary heart disease (CHD), the mechanism of traditional Chinese medicine (TCM) is still unclear because of its complex prescription rules. This study prospectively collected 715 prescriptions of TCM for the treatment of CHD. The characteristics of TCM in prescriptions were described and analyzed, and the rules of prescriptions were analyzed by using association rules. Frequency statistics showed that the high-frequency herbs with a frequency of more than 60% were Gan-cao, Huang-qi, Dang-gui, Chuan-xiong, Yan-hu-suo, and San-qi. The high-frequency herb combinations were summarized by using association rules. By using the method of the “Top N groups” to excavate the empirical prescriptions, the basic prescriptions for treating CHD were summarized. We named the intersection herbs of the basic prescriptions and the high frequency herbs as the core herbal prescription. To explore the possible mechanisms underlying the anti-CHD effect of the core herbal prescription, the bioactive components of core herbal prescription and their targets were screened out by using network pharmacology. Molecular docking was performed between the bioactive components and core targets. A total of 28 potential active ingredients and 5 core targets were identified for the treatment of CHD with core herbal prescription. The enrichment analysis results indicated that the mechanism of action mainly involved neuroactive ligand-receptor interaction and calcium signaling pathway. The commonly used herbal pairs for CHD with qi deficiency and blood stasis syndrome were Huang-qi and Dang-gui. The mechanism of action of common herbal pairs was also studied by network pharmacology. This study summarized the prescription rule of TCM in the treatment of CHD and may provide a new idea for the treatment of CHD.

Frequent pattern mining is the process of mining data in a set of items or some patterns from a largedatabase. The resulted frequent set data supports the minimum support threshold. A frequentpattern is a pattern that occurs frequently in a dataset. Association rule mining is defined as to findout association rules that satisfy the predefined minimum support and confidence from a given database. If an item set is said to be frequent, that item set supports the minimum support andconfidence. A Frequent item set should appear in all the transaction of that data base. Discoveringfrequent item sets play a very important role in mining association rules, sequence rules, web logmining and many other interesting patterns among complex data. Data stream is a real timecontinuous, ordered sequence of items. It is an uninterrupted flow of a long sequence of data. Somereal time examples of data stream data are sensor network data, telecommunication data,transactional data and scientific surveillances systems. These data produced trillions of updatesevery day. So it is very difficult to store the entire data. In that time some mining process is required.Data mining is the non-trivial process of identifying valid, original, potentially useful and ultimatelyunderstandable patterns in data. It is an extraction of the hidden predictive information from largedata base. There are lots of algorithms used to find out the frequent item set. In that Apriorialgorithm is the very first classical algorithm used to find the frequent item set. Apart from Apriori,lots of algorithms generated but they are similar to Apriori. They are based on prune and candidategeneration. It takes more memory and time to find out the frequent item set. In this paper, we havestudied about how the eclat algorithm is used in data streams to find out the frequent item sets.Eclat algorithm need not required candidate generation.