Frequent Mining Research Articles

Research on the Parallel Frequent Data Mining Strategy under the Cloud Computing Environment

Cloud computing is computing tasks distribution resources of a large number of computers in the subnet, to provide users with cheap and efficient computing power, storage capacity and service capabilities. Data mining is to find useful information in large data repository. Frequent flow of large amounts of data quickly and accurately find important basis for forecasting and decision, therefore, under the cloud computing environment parallelization frequent item data mining strategy to provide efficient solutions to store and analyze vast amounts of data has important theoretical significanceand application value.

Identifying relations between frequent patterns mined at two collaborative websites

In modern business world, very often two companies collaborate with each other for their mutual benefit in such a way that, the one starts a transaction and processes a part of it, then the other processes the remainder. Similarly, in cloud computing, as a means to avoid leakage of secret information, a company may use two independent cloud management domains to store separate partitions of its database. For many users in such application environment, it would be beneficial and important to discover the relations between frequent patterns mined at respective site, and share the frequent pattern relation identifiers. The frequent pattern relation mining should be conducted without disclosing any other private data to each other site. This paper identifies a new data mining problem called pattern relation mining, introduces a new computing model called IF-THEN computing to capture the problem, and proposes a privacy-preserving pattern relation mining algorithm called CPRM. Extensive experiments were conducted to demonstrate the effectiveness of CPRM.

High Utility Itemset Mining with SelectiveItem Replication

Incessant weighted thing sets speak to connections habitually holding in information in which things may weight in an unexpected way. Nonetheless, in a few settings, e.g., when the need is to minimize a certain expense capacity, finding uncommon information connections is more intriguing than mining regular ones Our technique works on a chart where vertices relate to incessant things and edges compare to successive thing arrangements of size two. Utility based information mining is another examination range inspired by a wide range of utility figures information mining techniques and focused at consolidating utility contemplations in information mining errands. Utility based information mining is another examination territory keen on a wide range of utility calculates information mining techniques and focused at consolidating utility contemplations in information mining undertakings. The UMining calculation is utilized to discover all high utility itemsets inside the given utility imperative limit. Quick Utility Frequent Mining, is a more exact and exceptionally late calculation. It takes both the utility and the bolster measure into thought. This strategy gives the itemsets that are both high utility as well as that may be, visit. Another idea is proposed for creating various types of itemsets specifically High utility and high successive itemsets (HUHF), High utility and low visit itemsets (HULF), Low utility and high regular itemsets (LUHF) and Low utility and low visit itemsets (LULF). These itemsets are produced utilizing the essential structure FP-Growth calculations.

Frequent Item Set Mining Using INC_MINE in Massive Online Analysis Frame Work

Frequent Pattern Mining is one of the major data mining techniques, which is exhaustively studied in the past decade. The technological advancements have resulted in huge data generation, having increased rate of data distribution. The generated data is called as a ‘data stream’. Data streams can be mined only by using sophisticated techniques. The paper aims at carrying out frequent pattern mining on data streams. Stream mining has great challenges due to high memory usage and computational costs. Massive online analysis frame work is a software environment used to perform frequent pattern mining using INC_MINE algorithm. The algorithm uses the method of closed frequent mining. The data sets used in the analysis are Electricity data set and Airline data set. The authors also generated their own data set, OUR-GENERATOR for the purpose of analysis and the results are found interesting. In the experiments five samples of instance sizes (10000, 15000, 25000, 35000, 50000) are used with varying minimum support and window sizes for determining frequent closed itemsets and semi frequent closed itemsets respectively. The present work establishes that association rule mining could be performed even in the case of data stream mining by INC_MINE algorithm by generating closed frequent itemsets which is first of its kind in the literature.

A General Framework Based on Composite Granules for Mining Association Rules

At present many algorithms for mining association rules have been proposed, but most of them are only suitable for discovering specific frequent itemsets from characteristic data sets on the appointed environments, namely, these algorithms are not general enough when mining association rules. In this paper, a general framework based on composite granules for mining association rules is proposed, which is a general data mining model without appointed restriction from frequent itemsets, data sets or mining environments and so on. An iterative method is efficiently applied to the general mining framework for discovering frequent itemsets, which adopts repartitioning frequent attributes to iteratively reconstruct the mixed radix information system for reducing a relational database. In order that the framework for discovering frequent itemsets has a generality, in discussing and establishing the general mining framework, this paper introduces a novel conception and data model, namely, a mixed radix information system is applied to describe a relational database, and a composite granules is used to build a specific relationship between an information system and a mixed radix information system, which can hold the same extension and simultaneously exist in two different information systems. The mixed radix information system can help the general framework to reduce information data and improve the performance of the framework for generating frequent itemsets. The composite granules model can create a relationship between an information granule and a digital information granule, and help the framework for computing the support to avoid reading the database repeatedly or using the complex data structure. Finally, a new taxonomy is presented to verify the generality and the high efficiency of the mining framework and all the experiments based on the taxonomy indicate that the general mining framework has the required generality, and the performance of the framework is better than these classical mining frameworks.

Identification of the sources of metals and arsenic in river sediments by multivariate analysis and geochemical approaches

Frequent mining activities and higher background values in soil have led to the contamination of the sediments of some rivers in southwest China by several metals and arsenic (As). This study combined multivariate analysis with geochemical approaches to differentiate mining activity from other sources, which may aid to evaluate the effectiveness of reducing mining release. Sixteen sediment samples were collected along the Yuan River, China. The total concentrations of lead (Pb), zinc (Zn), copper (Cu), cadmium (Cd), chromium (Cr), mercury (Hg), and As were measured by inductively coupled plasma-atomic emission spectrometer (ICP-AES). The Pb isotopic composition was measured using a thermal ionization mass spectrometer (TIMES). Both geochemical approaches and multivariate statistical analysis were used to identify the sources of these metals. The fractionation of Pb was determined through a Community Bureau of Reference (BCR) sequential extraction procedure to aid the identification of the sources. The concentrations and enrichment factors (EFs) of Pb, Zn, Cu, Cd, and As in the middle reach of the river were higher than those at the other sites, indicating anthropogenic sources. The factor analysis (FA) extracted “mining and smelting,” “mixture of anthropogenic and natural,” and “natural” factors. The Pb isotope composition of metal ores was similar (206Pb/207Pb 2.023) to that found in the sediments in the middle reach, indicating anthropogenic sources of mining activities. Compared with the narrow ranges of the δ34S ratios in the bedrock (+8.5 to +9.3 ‰) and the metal ores (−1.4 to +1.9 ‰), the sediment samples presented a relatively wide range of δ34S ratios from −2.6 to +9.2 ‰ with a mean of +2.6 ‰, which suggests a mixed composition. The BCR sequential extraction procedure revealed that the proportion of the extractable fraction in the sediments in the middle reach was higher than that in other sites, suggesting anthropogenic sources as the cause of contamination in the study area. Lead, Zn, Cu, Cd, Cr, Hg, and As are mainly derived from natural materials in the upstream region. In the middle reach, these elements are the result of anthropogenic activities, particularly activities associated with the mining industry. In the downstream region, the origin of these elements is considered to be a mixture of anthropogenic and natural sources. In addition to geochemical approaches and multivariate statistical analysis, the BCR sequential extraction method is an effective procedure for the identification of the anthropogenic sources of sediment-associated metals.

Research on Supply Chain Abnormal Event Detection Based on the RFID Technology

In a complex supply network composed of suppliers, manufactures and buyers, RFID tech is used to connect different items. RFID tech makes it possible to collect data in logistics with a low cost and it can optimize the traditional method more efficiently. A platform using EPCGlobal standards supplies the base environment for high-level applications, such as abnormal event detection in a supply chain, frequent path mining and the retail theft prediction. This paper studies abnormal events detection in the supply network. The data captured from the EPC information service is used to calculate a path and the machine learning method is adopted to cluster the path.

Study on Predicting Various Mining Techniques Using Weighted Itemsets

Pattern mining is one among datamining technique meant to reveal the existing patterns from data. Datamining is helpful for retrieving the hidden and most interesting patterns. Frequent mining can be obtained with and without candidate generation schemes. From the study of literature, complexity of data has been reduced by the utilization of Infrequent Pattern Mining method, which represents rare data correlations among data. The proposed approach is very helpful to handle the problem of predicting Infrequent and Weighted itemsets, which can be termed as Infrequent Weighted Itemset (IWI) mining problem. There are two methods have been used for handling IWI mining problem namely IWI and minimal IWI. To enhance the potential application such as medical science , biological datas by using IWI approach. How to extract the infrequent items from biological database depending on its weights and use the characteristics of the data how to develop some application. Experimental results show efficiency and effectiveness of the proposed algorithm. Index Terms: Clustering, Classification, Weighted Association rules and Infrequent pattern mining, Weighted support

Mining Maximal Adjacent Frequent Patterns from DNA Sequences using Location Information

rapid development of bioinformatics has resulted in the explosion of DNA sequence data which is characterized by large number of items. Studies have shown that biological functions are dictated by contagious portions of the DNA sequence. Finding contiguous frequent patterns from long data sequences such as DNA sequences is a particularly challenging task and can pave the way towards new breakthroughs. Apriori based techniques were among the first to be used in frequent contagious pattern mining. Later improved approaches like GSP, Prefix Span were also applied but the approaches required either large number of sequence scans, generated large number of candidates or required higher number of intermediate sequential patterns. In this paper an improvement of the positional based approach for contagious frequent pattern mining is DNA sequences is proposed. The proposed algorithm improves the existing positional based approach by introducing a new amalgamated sorting and joining technique which helps to reduce time and space complexity. The proposed approach outperforms traditional existing contiguous frequent mining approaches.

A DOMAIN-INDEPENDENT, TRANSFERABLE AND TIMELY ANALYSIS APPROACH TO ASSESS STUDENT COLLABORATION

Collaborative learning environments require intensive, regular and frequent analysis of the increasing amount of interaction data generated by students to assess that collaborative learning takes place. To support timely assessments that may benefit students and teachers the method of analysis must provide meaningful evaluations while the interactions take place. This research proposes machine learning-based techniques to infer the relationship between student collaboration and some quantitative domain-independent statistical indicators derived from large-scale evaluation analysis of student interactions. This paper (i) compares a set of metrics to identify the most suitable to assess student collaboration, (ii) reports on student evaluations of the metacognitive tools that display collaboration assessments from a new collaborative learning experience and (iii) extends previous findings to clarify modeling and usage issues. The advantages of the approach are: (1) it is based on domain-independent and generally observable features, (2) it provides regular and frequent data mining analysis with minimal teacher or student intervention, thereby supporting metacognition for the learners and corrective actions for the teachers, and (3) it can be easily transferred to other e-learning environments and include transferability features that are intended to facilitate its usage in other collaborative and social learning tools.

Network Threat Characterization in Multiple Intrusion Perspectives using Data Mining Technique

For effective security incidence response on the network, a reputa ble approach must be in placeat both protected and unprotected region of the network. Thisis because compromise in the demilitarized zone could be precursor to threat inside the network. The improv ed complexity of attacks in present times and vulnerability of system are motivationsfor this work. Past and present approachesto intrusion detection and prevention have neglected victim and attackerproperties despite the fact that for intrusion to occur, an overt act by an attacker and a manifestation, observable by the intended victim, which results from that act are required. Therefore, this paper presents athreat characterization model forattacks from the victim and the attacker perspective of intrusionusing data mining technique. The data mining technique combines Frequent Temporal Sequence Assoc iation Mining and Fuzzy Logic. AprioriAssociation Mining algorithm was used to mine temporal rule patterns from alertsequences while Fuzzy Control Systemwas used to rate exploits. The results of the experiment showthat accurate threat characterization inmultiple intrusion perspectives could be actualized using Fuzzy Association Mining . Also, the results proved that sequence of exploits could be used to rate threat and are motivated by victim properties and attacker objectives.

A data mining approach to evolutionary optimisation of noisy multi-objective problems

Many real world optimisation problems have opposing objective functions which are subjected to the influence of noise. Noise in the objective functions can adversely affect the stability, performance and convergence of evolutionary optimisers. This article proposes a Bayesian frequent data mining (DM) approach to identify optimal regions to guide the population amidst the presence of noise. The aggregated information provided by all the solutions helped to average out the effects of noise. This article proposes a DM crossover operator to make use of the rules mined. After implementation of this operator, a better convergence to the true Pareto front is achieved at the expense of the diversity of the solution. Consequently, an ExtremalExploration operator will be proposed in the later part of this article to help curb the loss in diversity caused by the DM operator. The result is a more directive search with a faster convergence rate. The search is effective in decision space where the Pareto set is in a tight cluster. A further investigation of the performance of the proposed algorithm in noisy and noiseless environment will also be studied with respect to non-convexity, discontinuity, multi-modality and uniformity. The proposed algorithm is evaluated on ZDT and other benchmarks problems. The results of the simulations indicate that the proposed method is effective in handling noise and is competitive against the other noise tolerant algorithms.

Data Mining Approach Using Apriori Algorithm: The Review

Mining frequent itemsets is one of the most investigated fields in data mining. It is a fundamental and crucial task. In data mining approach, the quantitative attributes should be appropriately dealt with as well as the Boolean attributes. Apriori Algorithm is one the best methods to extract the frequent mining Data Set. This paper gives us a brief review of apriori algorithm along with its uses to various fields and with various algorithms.

Comparative Study of Frequent Pattern Mining Techniques

Frequent pattern mining has been a focused theme in data mining research. Lots of techniques have been proposed to improve the performance of frequent pattern mining algorithms. This paper presents review of different frequent mining techniques. With each technique, we have provided brief description of the technique. At the end, we compared different frequent pattern mining techniques.

Comparative Analysis of Various Approaches Used in Frequent Pattern Mining

Frequent pattern mining has become an important data mining task and has been a focused theme in data mining research. Frequent patterns are patterns that appear in a data set frequently. Frequent pattern mining searches for recurring relationship in a given data set. Various techniques have been proposed to improve the performance of frequent pattern mining algorithms. This paper presents review of different frequent mining techniques including apriori based algorithms, partition based algorithms, DFS and hybrid algorithms, pattern based algorithms, SQL based algorithms and Incremental apriori based algorithms. A brief description of each technique has been provided. In the last, different frequent pattern mining techniques are compared based on various parameters of importance. Experimental results show that FP- Tree based approach achieves better performance.

Maximal Frequent Item Sequences Mining of Datasets with few Attributes and Large Instances

This work proposes a new fast algorithm named MFISS-FG (maximal frequent item sequence sets fast generating) finding maximal frequent item sequences from relational database. It adapts to datasets with few attributes and large instances. Itemset is defined as IS (item sequence) for mining. Two lists called ISL (Item Sequence List) and FISL (Frequent Item Sequence List) are created by scanning database once for dividing n-IS into two categories depending on whether the IS to achieve minimum support number (n is the number of attributes). SIS (Sub item sequences) whose n-superset is in ISL are generated by recursion to make sure that each k-SIS appeared before its (k+1)-superset (k range from 1 to n-1). As current k-SIS being joined to FISL, its (k-1)-SIS are pruned (k range from 2 to n-1). At last, all SISs whose n-superset is in FISL are pruned from FISL to hold all maximal frequent item sequences. We compare our MFISS-FG and FP-Growth by a set of time-consuming experiments to prove the superiority of MFISS-FG both not only with increasing datasets but also with changing mini-support.

Maximal Frequent Itemsets in Data Stream Mining Based on Orderly-Compound Policy

Mining maximal frequent itemsets get the advantage of a relatively small number of itemsets. Compared to mining frequent itemsets and mining frequent closed itemsets, such algorithm has higher time and space efficiency. According to the features of data streams and combined sliding window, a new algorithm E-FPMFI which is based on orderly-compound policy for mining maximal frequent itemsets in data stream is proposed. The algorithm based on basic window updates information from data stream flow fragment and scans the stream only once to gain and store it in frequent itemsets list. The algorithm construct FP-tree, then compress orderly FP-tree by merging nodes which has equal minsup in same branch, also uses subset mix pruning technique, avoid superset checking. The experimental results show the algorithm has higher time, space efficiency and good scalability.

An Optimized LCP Table Based Algorithm for Frequent String Mining

Given m databases D1,...,Dm of strings, the purpose of the frequent string mining is to find all strings that fulfill certain constraints of all string databases. In this paper, a useful data structure is proposed to construct suffix and LCP table which can reduce the total space consumption of string mining efficiently. We demonstrate the use of this data structure by optimizing the algorithm proposed by A.Kügel et al [7] and present the improved algorithm. It is achieved that the space consumption in our algorithm is proportional to the length of the largest string of all databases. A set of comprehensive performance experiments shows that the processing rate is enhanced because amount of items are reduced in new data structure.

An efficient incremental mining algorithm-QSD

The generation of frequent itemsets is an essential and time-consuming step in mining association rules. Most of the studies adopt the Apriori-based approach, which has great effort in generating candidate itemsets and needs multiple database accesses. Recent studies indicate that FP-tree approach has been utilized to avoid the generation of candidate itemsets and scan transaction database only twice, but they work with more complicated data structure. Besides, it needs to adjust the structure of FP-tree when it applied to incremental mining application. It is necessary to adjust the position of an item upward or downward in the structure of FP-tree when a new transaction increases or decreases the accumulation of the item. The process of the adjustment of the structure of FP-tree is the bottlenecks of the FP-tree in incremental mining application. Therefore, algorithms for efficient mining of frequent patterns are in urgent demand. This paper aims to improve both time and space efficiency in mining frequent itemsets and incremental mining application. We propose a novel QSD (Quick Simple Decomposition) algorithm using simple decompose principle which derived from minimal heap tree, we can discover the frequent itemsets quickly under one database scan. Meanwhile, QSD algorithm doesn't need to scan database and reconstruct data structure again when database is updated or minimum support is varied. It can be applied to on-line incremental mining applications without any modification. Comprehensive experiments have been conducted to assess the performance of the proposed algorithm. The experimental results show that the QSD algorithm outperforms previous algorithms.

Frequent mining of subgraph structures

In recent years, frequent graph pattern mining has raised great interest in the data mining community and has been applied to bioinformatics, web exploration, etc. The difficulty is that the search of possible subgraph patterns is combinatorially explosive. Many researchers proposed methods to resolve this problem, such as FSG, gSpan, and CloseGraph. But we discovered that the result set generated by these methods contains many repetitive structures. In this article, we propose a new method called Frequent Structure Mining Algorithm (FSMA) to find frequent structure in graph database, and prove corresponding theoretics to guarantee the accuracy of FSMA. The power of FSMA comes from the resolution of finding different structures and the storage of the graph set; it guarantees the completeness of mining results.