Dissertations / Theses on the topic 'MINING ASSOCIATION RULES'
To see the other types of publications on this topic, follow the link: MINING ASSOCIATION RULES.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'MINING ASSOCIATION RULES.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Cai, Chun Hing. "Mining association rules with weighted items." Hong Kong : Chinese University of Hong Kong, 1998. http://www.cse.cuhk.edu.hk/%7Ekdd/assoc%5Frule/thesis%5Fchcai.pdf.
Full textAbstract:
Thesis (M. Phil.)--Chinese University of Hong Kong, 1998.Description based on contents viewed Mar. 13, 2007; title from title screen. Includes bibliographical references (p. 99-103). Also available in print.
2
Zhou, Zequn. "Maintaining incremental data mining association rules." Thesis, National Library of Canada = Bibliothèque nationale du Canada, 2001. http://www.collectionscanada.ca/obj/s4/f2/dsk3/ftp05/MQ62311.pdf.
Full text
3
Goulbourne, Graham. "Tree algorithms for mining association rules." Thesis, University of Liverpool, 2001. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.250218.
Full textAbstract:
With the increasing reliability of digital communication, the falling cost of hardware and increased computational power, the gathering and storage of data has become easier than at any other time in history. Commercial and public agencies are able to hold extensive records about all aspects of their operations. Witness the proliferation of point of sale (POS) transaction recording within retailing, digital storage of census data and computerized hospital records. Whilst the gathering of such data has uses in terms of answering specific queries and allowing visulisation of certain trends the volumes of data can hide significant patterns that would be impossible to locate manually. These patterns, once found, could provide an insight into customer behviour, demographic shifts and patient diagnosis hitherto unseen and unexpected. Remaining competitive in a modem business environment, or delivering services in a timely and cost effective manner for public services is a crucial part of modem economics. Analysis of the data held by an organisaton, by a system that "learns" can allow predictions to be made based on historical evidence. Users may guide the process but essentially the software is exploring the data unaided. The research described within this thesis develops current ideas regarding the exploration of large data volumes. Particular areas of research are the reduction of the search space within the dataset and the generation of rules which are deduced from the patterns within the data. These issues are discussed within an experimental framework which extracts information from binary data.
4
Koh, Yun Sing, and n/a. "Generating sporadic association rules." University of Otago. Department of Computer Science, 2007. http://adt.otago.ac.nz./public/adt-NZDU20070711.115758.
Full textAbstract:
Association rule mining is an essential part of data mining, which tries to discover associations, relationships, or correlations among sets of items. As it was initially proposed for market basket analysis, most of the previous research focuses on generating frequent patterns. This thesis focuses on finding infrequent patterns, which we call sporadic rules. They represent rare itemsets that are scattered sporadically throughout the database but with high confidence of occurring together. As sporadic rules have low support the minabssup (minimum absolute support) measure was proposed to filter out any rules with low support whose occurrence is indistinguishable from that of coincidence. There are two classes of sporadic rules: perfectly sporadic and imperfectly sporadic rules. Apriori-Inverse was then proposed for perfectly sporadic rule generation. It uses a maximum support threshold and user-defined minimum confidence threshold. This method is designed to find itemsets which consist only of items falling below a maximum support threshold. However imperfectly sporadic rules may contain items with a frequency of occurrence over the maximum support threshold. To look for these rules, variations of Apriori-Inverse, namely Fixed Threshold, Adaptive Threshold, and Hill Climbing, were proposed. However these extensions are heuristic. Thus the MIISR algorithm was proposed to find imperfectly sporadic rules using item constraints, which capture rules with a single-item consequent below the maximum support threshold. A comprehensive evaluation of sporadic rules and current interestingness measures was carried out. Our investigation suggests that current interestingness measures are not suitable for detecting sporadic rules.
5
Pray, Keith A. "Apriori Sets And Sequences: Mining Association Rules from Time Sequence Attributes." Link to electronic thesis, 2004. http://www.wpi.edu/Pubs/ETD/Available/etd-0506104-150831/.
Full textAbstract:
Thesis (M.S.) -- Worcester Polytechnic Institute.Keywords: mining complex data; temporal association rules; computer system performance; stock market analysis; sleep disorder data. Includes bibliographical references (p. 79-85).
6
王漣 and Lian Wang. "A study on quantitative association rules." Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 1999. http://hub.hku.hk/bib/B31223588.
Full text
7
Wang, Lian. "A study on quantitative association rules /." Hong Kong : University of Hong Kong, 1999. http://sunzi.lib.hku.hk/hkuto/record.jsp?B2118561X.
Full text
8
Zhu, Hua. "On-line analytical mining of association rules." Thesis, National Library of Canada = Bibliothèque nationale du Canada, 1998. http://www.collectionscanada.ca/obj/s4/f2/dsk2/ftp01/MQ37678.pdf.
Full text
9
Wu, Jingtong. "Interpretation of association rules with multi-tier granule mining." Thesis, Queensland University of Technology, 2014. https://eprints.qut.edu.au/71455/1/Jing_Wu_Thesis.pdf.
Full textAbstract:
This study was a step forward to improve the performance for discovering useful knowledge – especially, association rules in this study – in databases. The thesis proposed an approach to use granules instead of patterns to represent knowledge implicitly contained in relational databases; and multi-tier structure to interpret association rules in terms of granules. Association mappings were proposed for the construction of multi-tier structure. With these tools, association rules can be quickly assessed and meaningless association rules can be justified according to the association mappings. The experimental results indicated that the proposed approach is promising.
10
Delpisheh, Elnaz, and University of Lethbridge Faculty of Arts and Science. "Two new approaches to evaluate association rules." Thesis, Lethbridge, Alta. : University of Lethbridge, Dept. of Mathematics and Computer Science, c2010, 2010. http://hdl.handle.net/10133/2530.
Full textAbstract:
Data mining aims to discover interesting and unknown patterns in large-volume data. Association rule mining is one of the major data mining tasks, which attempts to find inherent relationships among data items in an application domain, such as supermarket basket analysis. An essential post-process in an association rule mining task is the evaluation of association rules by measures for their interestingness. Different interestingness measures have been proposed and studied. Given an association rule mining task, measures are assessed against a set of user-specified properties. However, in practice, given the subjectivity and inconsistencies in property specifications, it is a non-trivial task to make appropriate measure selections. In this work, we propose two novel approaches to assess interestingness measures. Our first approach utilizes the analytic hierarchy process to capture quantitatively domain-dependent requirements on properties, which are later used in assessing measures. This approach not only eliminates any inconsistencies in an end user’s property specifications through consistency checking but also is invariant to the number of association rules. Our
second approach dynamically evaluates association rules according to a composite and
collective effect of multiple measures. It interactively snapshots the end user’s domain-
dependent requirements in evaluating association rules. In essence, our approach uses
neural networks along with back-propagation learning to capture the relative importance
of measures in evaluating association rules. Case studies and simulations have been conducted to show the effectiveness of our two approaches.viii, 85 leaves : ill. ; 29 cm
11
魯建江 and Kin-kong Loo. "Efficient mining of association rules using conjectural information." Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 2001. http://hub.hku.hk/bib/B31224878.
Full text
12
Loo, Kin-kong. "Efficient mining of association rules using conjectural information." Hong Kong : University of Hong Kong, 2001. http://sunzi.lib.hku.hk/hkuto/record.jsp?B22505544.
Full text
13
Palanisamy, Senthil Kumar. "Association rule based classification." Link to electronic thesis, 2006. http://www.wpi.edu/Pubs/ETD/Available/etd-050306-131517/.
Full textAbstract:
Thesis (M.S.)--Worcester Polytechnic Institute.Keywords: Itemset Pruning, Association Rules, Adaptive Minimal Support, Associative Classification, Classification. Includes bibliographical references (p.70-74).
14
Yang, Yuping. "Theory and mining of association rules over large databases /." The Ohio State University, 2000. http://rave.ohiolink.edu/etdc/view?acc_num=osu1488194825668668.
Full text
15
Zhang, Hongen. "Mining and visualization of association rules over relational DBMSs." [Florida] : State University System of Florida, 2000. http://etd.fcla.edu/etd/uf/2000/ana7033/Etd.pdf.
Full textAbstract:
Thesis (M.S.)--University of Florida, 2000.Title from first page of PDF file. Document formatted into pages; contains xiii, 100 p.; also contains graphics. Vita. Includes bibliographical references (p. 97-99).
16
Xiang, Lan. "Interesting Association Rules Mining Based on Improved Rarity Algorithm." Thesis, Mittuniversitetet, Avdelningen för informationssystem och -teknologi, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:miun:diva-35320.
Full textAbstract:
With the rapid development of science and technology, our society has been in the big data era. In human activities, we produce a lot of data in every second and every minute, what contain much information. Then, how to select the useful information from those complicated data is a significant issue. So the association rules mining, a technique of mining patterns or associations between itemsets, comes into being. And this technique aims to find some important associations in data to get useful knowledge. Nowadays, most scholars at home and abroad focus on the frequent pattern mining. However, it is undeniable that the rare pattern mining also plays an important role in many areas, such as the medical, financial, and scientific field. Comparing with frequent pattern mining, studying rare pattern mining is more valuable, because it tends to find unknown, unexpected, and more interesting rules. But the study of rare pattern mining is little difficult because of the scarcity of data used for verifying rules. In the frequent pattern mining, there are two general algorithms of discovering frequent itemsets, i.e., Apriori, the earliest algorithm which is proposed by R.Agrawal in 1994, and FP-Tree, the improved algorithm which reduced the time complexity. And in rare pattern mining, there are also two algorithms, Arima and Rarity, what are similar to Apriori and FP-Tree algorithms, but they still exist some problems, for example, Arima is time-consuming because of repeatedly scanning the large database, and Rarity is space-consuming because of the establishment of the full-combination tree. Therefore, based on the Rarity algorithm, this report presents an improved method to efficiently discover interesting association rules among rare itemsets and aims to get a balance between time and space. It is a top-down strategy which uses the graph structure to indicate all combinations of existing items, defines pattern matrix to record itemsets, and combines the hash table to accelerate calculation process. This method decreases both the time cost and the space cost when comparing with Arima, and reduces the space waste to solve the problem of Rarity, but its searching time of mining rare itemsets is more than Rarity, and we verified the feasibility of this algorithm only on abstract and small databases. Thus in the future, on the one hand, we will continue improving our method to explore how to decrease the searching time in the process and adjust the hash function to optimize the space utilization. And on the other hand, we will apply our method to actual large databases, such as the clinical database of the diabetic patients to mine association rules in diabetic complications.
17
李守敦 and Sau-dan Lee. "Maintenance of association rules in large databases." Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 1997. http://hub.hku.hk/bib/B31215531.
Full text
18
Chen, Wei-Ren, and 陳威任. "Mining Utility Association Rules." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/04865121871313091524.
Full textAbstract:
碩士銘傳大學
資訊工程學系碩士班
103
Mining Association Rules can find which products would be purchased by the customer when a customer has bought some products, and we can use association rules to recommend products for customers. Mining High Utility Itemset is to find the combinations of products which could bring high profit to us. However, High Utility Itemset only tells us which products bring high profit but not increase profit when we recommend other product to customer. Therefore, we propose definitions and algorithm of Mining Utility Association Rules to find which product to recommend and to bring us more benefit than the original high utility itemsets. We will clearly know which product should be recommended to customer bring more profit to us with Utility Association Rules.
19
Li, Li-Ya, and 李立雅. "Inter-sequence Association Rules Mining." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/07341599579887641679.
Full textAbstract:
碩士國立臺灣大學
資訊管理研究所
91
There are many algorithms proposed to find sequential patterns in sequence databases where each transaction contains one sequence. Previously proposed algorithms treat each sequence as an independent one. This kind of mining belongs to intra-transaction sequential patterns mining. In this paper, we propose an algorithm, ProbSif, to mine inter-sequence association rules. Our proposed algorithm consists of three phases. First, we find all large intra-sequence patterns. For each large pattern found, all the time points at which the pattern occurs are recorded in a time point list. Second, those time point lists are hashed into L-buckets. Third, we use a level-wise candidate generation-and-test method to generate candidate patterns across different sequences and check if a candidate is large. Once we generate a candidate, we count its support by reading relevant time point lists from L-buckets. By using the L-buckets, our proposed algorithm requires fewer database scans than the Apriori-like approach. Therefore, our proposed algorithm is more efficient. The experimental results show that our proposed algorithm outperforms the Apriori-like approach by several orders of magnitude.
20
Chang, Paul C. M., and 張仲銘. "Mining Association Rules by Sorts." Thesis, 1998. http://ndltd.ncl.edu.tw/handle/27186430188696978772.
Full textAbstract:
碩士國立清華大學
資訊工程學系
86
In this thesis, we use the knowledge about the sorts of items and transactions to discover association rules among items in a market transaction database. It is natural to divide items into sorts: milk and bread belong to the sort of food while gloves and hats pertain to the sort of clothing. We sort each transaction according to the sorts of items contained by this transaction. Then each sort of transactions will form a subset of the entire database. To discover the association rules within and between these subsets, two kinds of support-constraint models with the corresponding algorithms are proposed. We claim that such models not only enrich the semantics of rules compared with the inceptive work but also emphasize the customer buying patterns for both intra-sort and inter-sort merchandise. The constraint needed when generating rules based on sorts of items is also discussed. The experiments evaluate the performance of these algorithms on synthetical databases of different inter- sort patterns.
21
Su, Wei-Tu, and 蘇威圖. "Mining Multidimensional Intertransaction Association Rules." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/54923326716084839179.
Full textAbstract:
碩士國立臺灣大學
資訊管理研究所
90
Traditionally, association rule data mining almost focuses on finding the associations among items within the same transaction. In this thesis, we explore “Multidimensional Intertrnasaction Association Rules”, which tries to find the association rule from different transactions and extend to multidimensional space. We propose the E-Partition algorithm and use the Grid File as our data structure to find the large itemsets in the database. Besides, we propose the E-DELTA algorithm to deal with the incremental data mining. The experiment shows that the E-Partition algorithm performs better than the E-Apriori algorithm. Also, the algorithm using the Grid File has better efficiency than that scanning database does.
22
Chien, Peng Wang, and 王建鵬. "Find the General Rule of Data Mining Association Rules." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/08735074145658888662.
Full textAbstract:
碩士萬能科技大學
資訊管理研究所
99
At present the application of association rule mining and research, to exchange products generated discussion targeted mostly clustered, and in the exploration process and output that, there is no a general rule of representation, usually in a unique way or the text description . This study proposes a concept of transactions by participants in the association rule mining as an object. For association rule mining applications more flexible, to entities associated with the set methodology for the extension of a graphical representation, so that regardless of the implementation of the method, the can be simple and clear expression, and association rule mining to fully describe the various restrictions , regardless of entity-relationship structure, star structure, snow structure, can be described as a class can be summarized, and describe the relationship between different induction levels. Another object via the specified mining, exploration using different trading partners, meaning more like mining.
23
Yang, Chian-Yi, and 楊千儀. "Mining High Utility Quantitative Association Rules." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/2jdtaf.
Full textAbstract:
碩士銘傳大學
資訊傳播工程學系碩士班
94
Mining weighted association rules consider the importance of items in a large transaction database. Mining quantitative association rules find most quantitative itemsets , which are purchased frequently, and relate with them from a large transaction database. However, weighted association rules didn’t consider the items which their quantities, and quantitative association rules didn’t consider the items which their weighted. Economics mention influence that quantities affect the cost; and high prices are not necessarily to make a profit, that proves, if only consider weighted or quantitative, it’s must not enough. This paper will consider both weighted and quantitative, and find out useful rules for policymaker. We will weight of items multiply quantitative of items, it’s mean utility, and we want to find high utility association rules that these items reach to the utility threshold. Our methods don’t produce candidates and just scan once database to produce about sub-database, then we use these sub-database to find profitable association rules.
24
Yi-Ling, Chen. "Mining Spatial Association Rules in Image." 2005. http://www.cetd.com.tw/ec/thesisdetail.aspx?etdun=U0001-0907200516580400.
Full text
25
Su, Po-Ta, and 蘇伯達. "Mining Association Rules by Ant System." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/51532702116386182507.
Full textAbstract:
碩士國立清華大學
工業工程與工程管理學系
90
Mining association rules is to find relations among large amount of data so that the pattern of the dataset can be discovered. Many companies use association rules to find the relations among different items to improve their service quality of customers or enlarge their marketplace. Recently, many algorithms have been developed that only consider either non-quantitative data or quantitative data. However, in reality, most data we collected are mixed in types. Since Ant System allows to consider both of data types and has advantages of being efficient in filtering the unobvious association rules to reduce the unnecessary outputs and ease of making judgment to improve the performance, therefore, in this study, we adopted the technique and concept of Ant System to develop association rules. The developed algorithm is supported by theoretical evidence, and comparative studies are provided for evaluation.
26
"Mining association rules with weighted items." 1998. http://library.cuhk.edu.hk/record=b5889513.
Full textAbstract:
by Cai, Chun Hing.Thesis (M.Phil.)--Chinese University of Hong Kong, 1998.
Includes bibliographical references (leaves 109-114).
Abstract also in Chinese.
Acknowledgments --- p.ii
Abstract --- p.iii
Chapter 1 --- Introduction --- p.1
Chapter 1.1 --- Main Categories in Data Mining --- p.1
Chapter 1.2 --- Motivation --- p.3
Chapter 1.3 --- Problem Definition --- p.4
Chapter 1.4 --- Experimental Setup --- p.5
Chapter 1.5 --- Outline of the thesis --- p.6
Chapter 2 --- Literature Survey on Data Mining --- p.8
Chapter 2.1 --- Statistical Approach --- p.8
Chapter 2.1.1 --- Statistical Modeling --- p.9
Chapter 2.1.2 --- Hypothesis testing --- p.10
Chapter 2.1.3 --- Robustness and Outliers --- p.11
Chapter 2.1.4 --- Sampling --- p.12
Chapter 2.1.5 --- Correlation --- p.15
Chapter 2.1.6 --- Quality Control --- p.16
Chapter 2.2 --- Artificial Intelligence Approach --- p.18
Chapter 2.2.1 --- Bayesian Network --- p.19
Chapter 2.2.2 --- Decision Tree Approach --- p.20
Chapter 2.2.3 --- Rough Set Approach --- p.21
Chapter 2.3 --- Database-oriented Approach --- p.23
Chapter 2.3.1 --- Characteristic and Classification Rules --- p.23
Chapter 2.3.2 --- Association Rules --- p.24
Chapter 3 --- Background --- p.27
Chapter 3.1 --- Iterative Procedure: Apriori Gen --- p.27
Chapter 3.1.1 --- Binary association rules --- p.27
Chapter 3.1.2 --- Apriori Gen --- p.29
Chapter 3.1.3 --- Closure Properties --- p.30
Chapter 3.2 --- Introduction of Weights --- p.31
Chapter 3.2.1 --- Motivation --- p.31
Chapter 3.3 --- Summary --- p.32
Chapter 4 --- Mining weighted binary association rules --- p.33
Chapter 4.1 --- Introduction of binary weighted association rules --- p.33
Chapter 4.2 --- Weighted Binary Association Rules --- p.34
Chapter 4.2.1 --- Introduction --- p.34
Chapter 4.2.2 --- Motivation behind weights and counts --- p.36
Chapter 4.2.3 --- K-support bounds --- p.37
Chapter 4.2.4 --- Algorithm for Mining Weighted Association Rules --- p.38
Chapter 4.3 --- Mining Normalized Weighted association rules --- p.43
Chapter 4.3.1 --- Another approach for normalized weighted case --- p.45
Chapter 4.3.2 --- Algorithm for Mining Normalized Weighted Association Rules --- p.46
Chapter 4.4 --- Performance Study --- p.49
Chapter 4.4.1 --- Performance Evaluation on the Synthetic Database --- p.49
Chapter 4.4.2 --- Performance Evaluation on the Real Database --- p.58
Chapter 4.5 --- Discussion --- p.65
Chapter 4.6 --- Summary --- p.66
Chapter 5 --- Mining Fuzzy Weighted Association Rules --- p.67
Chapter 5.1 --- Introduction to the Fuzzy Rules --- p.67
Chapter 5.2 --- Weighted Fuzzy Association Rules --- p.69
Chapter 5.2.1 --- Problem Definition --- p.69
Chapter 5.2.2 --- Introduction of Weights --- p.71
Chapter 5.2.3 --- K-bound --- p.73
Chapter 5.2.4 --- Algorithm for Mining Fuzzy Association Rules for Weighted Items --- p.74
Chapter 5.3 --- Performance Evaluation --- p.77
Chapter 5.3.1 --- Performance of the algorithm --- p.77
Chapter 5.3.2 --- Comparison of unweighted and weighted case --- p.79
Chapter 5.4 --- Note on the implementation details --- p.81
Chapter 5.5 --- Summary --- p.81
Chapter 6 --- Mining weighted association rules with sampling --- p.83
Chapter 6.1 --- Introduction --- p.83
Chapter 6.2 --- Sampling Procedures --- p.84
Chapter 6.2.1 --- Sampling technique --- p.84
Chapter 6.2.2 --- Algorithm for Mining Weighted Association Rules with Sampling --- p.86
Chapter 6.3 --- Performance Study --- p.88
Chapter 6.4 --- Discussion --- p.91
Chapter 6.5 --- Summary --- p.91
Chapter 7 --- Database Maintenance with Quality Control method --- p.92
Chapter 7.1 --- Introduction --- p.92
Chapter 7.1.1 --- Motivation of using the quality control method --- p.93
Chapter 7.2 --- Quality Control Method --- p.94
Chapter 7.2.1 --- Motivation of using Mil. Std. 105D --- p.95
Chapter 7.2.2 --- Military Standard 105D Procedure [12] --- p.95
Chapter 7.3 --- Mapping the Database Maintenance to the Quality Control --- p.96
Chapter 7.3.1 --- Algorithm for Database Maintenance --- p.98
Chapter 7.4 --- Performance Evaluation --- p.102
Chapter 7.5 --- Discussion --- p.104
Chapter 7.6 --- Summary --- p.105
Chapter 8 --- Conclusion and Future Work --- p.106
Chapter 8.1 --- Summary of the Thesis --- p.106
Chapter 8.2 --- Conclusions --- p.107
Chapter 8.3 --- Future Work --- p.108
Bibliography --- p.108
Appendix --- p.115
Chapter A --- Generating a random number --- p.115
Chapter B --- Hypergeometric distribution --- p.116
Chapter C --- Quality control tables --- p.117
Chapter D --- Rules extracted from the database --- p.120
27
Chen, Yi-Ling, and 陳奕伶. "Mining Spatial Association Rules in Image." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/39410529091384817730.
Full textAbstract:
碩士國立臺灣大學
電機工程學研究所
93
In this paper, we integrate data mining with image processing for discovering spatial relationships in images. We present an image mining framework, Spatial Association Rulemining (SAR), to mine spatial associations located in specific locations of images. A rule in the SAR refers to the occurrences of image content in a pair of spatial locations. The proposed approach is applied to mine color spatial association rules (color-SAR) in landscape scene images so as to demonstrate that the spatial association rules is able to the application of image classification. Our experimental results show that the classification accuracy of 86% can be achieved by the rule-based classifier.
28
Huang, Minghua, and 黃明華. "Algorithms for Parallel Association Rules Mining." Thesis, 1999. http://ndltd.ncl.edu.tw/handle/88224762660139352543.
Full textAbstract:
碩士國立臺灣科技大學
管理研究所資訊管理學程
87
Mining association rules is an important task. Many parallel algorithms have been proposed to expedite the execution of the mining process. In this thesis, we propose a parallel algorithm called ''PBSM'' for shared-disk environments, and implement the PBSM algorithm on an nCUBE parallel computer. In the PBSM algorithm, mining process is divided into two steps. In the first step, multiple processors are used to generate frequent itemsets. Then, in the second phase, a chosen processor is used to generate the related association rules. Through boolean-based table operations, the PBSM algorithm needs not generate candidate itemsets─which constitute the major part of execution time in the previous Apriori-based mining algorithms. Further-more, in the PBSM algorithm, each processor works independently in generating frequent itemsets. There is no need to send messages for itemsets, supports or counts between processors. As a result, our PBSM algorithm shows a superb performance compared to the existing parallel mining algorithms.
29
Zhen, Hao, and 振昊. "DPARM: Differential Privacy Association Rules Mining." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/xqj7yw.
Full textAbstract:
碩士國立臺灣大學
電機工程學研究所
107
In contemporary society, the rapid expansion of data volume has driven the development of data analysis techniques, which makes decision automation possible. Association analysis is an important task in data analysis. The goal is to find all co-occurrence relationships from the transactional dataset, i.e. frequent itemsets or confident association rules. An association rule consists of two parts, the antecedent and the consequent, which means that if the antecedent occurs then the consequent is also possible to happen. Confident association rules are those association rules with larger possibility, which can help people better discover patterns and develop corresponding strategies. The process of data analysis can be highly summarized as a set of queries, where each query is a real-valued function of the dataset. However, without any restriction and protection, accessing the dataset to answer the queries may lead to the disclosure of individual privacy. Therefore, techniques for privacy-preserving data analysis has received increasing attention. People are eager to find a strong, mathematically rigorous, and socio-cognitive-conform definition of privacy. Differential privacy is such a privacy definition that manages and quantifies the privacy risks faced by individuals in data analysis through the parameter called the privacy level. In general, differential privacy can be achieved by adding delicate noise to the query results. In this thesis, we focus on differential privacy association rules mining with multiple support thresholds, and solve the challenges existing in the state-of-art works. We propose and implement the DPARM algorithm, which uses multiple support thresholds to reduce the number of candidate itemsets while reflecting the real nature of the items, and uses random truncation and uniform partition to lower the dimensionality of the dataset. Both of these are helpful to reduce the sensitivity of the queries, thereby reducing the scale of the required noise and improving the utility of the mining results. We also stabilize the noise scale by adaptively allocating the privacy levels, and bound the overall privacy loss. In addition, we prove that the DPARM algorithm satisfies ex post differential privacy, and verify the utility of the DPARM algorithm through a series of experiments.
30
Wu, Chieh-Ming, and 吳界明. "Data mining for generalized association rules and privacy preservingData mining for generalized association rules and privacy preserving." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/85543535661382633122.
Full textAbstract:
博士國立雲林科技大學
工程科技研究所博士班
99
Data mining is an analysis method used to extract the unknown and latent information that hides in large dataset which has usable information. In the last few years the data mining model and method have long-term progress and the association rule mining is most often applied. The association rule research focus on discussion how to discover single level association rule effectiveness in the large dataset. In the recent years more and more researchers start to study the problem of multiple level association rules that was advantageous in the knowledge economy modernized society. In accordance to the enterprise, it must utilize nimbly the more deeply and more detailed association rules to assist the superintendent to complete policy-making in the short time. For reach the above objective, this study proposed an efficient data structure, Frequent Closed Enumerable Table (FCET), to speed the generalized association rules mining. In the other aspect, as a result of enterprise globalization acceleration, many sensitive individual information collection, processing and application involve to the individual privacy protection law. In addition, databases managed by enterprises also largely grow up. The databases store many individual sensitive material and corporation secret information. If the database suffers non-suitable access, it leads the security problem. Moreover, it causes the company secret restricted data and the individual material to be disclosed. Once the problem is not careful processed, it would possibly reduce the competitiveness of enterprise. This study proposes an effective data structure which considers the privacy preserving in the mining process. In addition, it carries on the complete discussion from data mining and privacy the preserving related question. A greedy algorithm which considers the hiding cost was proposed here. The algorithm includes the sanitized procedure and exposed procedure protection of mechanism. Not only privacy preserving for public content but also useful information extraction are guarantee to reach. Moreover, after the sanitized processing, it achieves privacy preserving and knowledge extracting balanced effectively.
31
Wang, Wei-Tse, and 王威澤. "A native XML database association rules mining method and a database compression approach using association rules mining." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/47594741243843634353.
Full textAbstract:
碩士朝陽科技大學
資訊管理系碩士班
91
With the advancement of technology and popularity of applications in enterprises’ information system, greater and greater amount of data is generated everyday. To properly store and access these data, database applications have come into play and become crucial. The main task of data mining is to help enterprises make their decisions by extracting useful information from the large amount of complicated data storage for reference, so this is why data mining has been recently paid more attention than ever. Also, more storage media for data is required for the increasing amount of data. For unlimited needs of increasing amount of data, it will be wise to provide an efficient data compression technique to reduce the cost. The thesis proposes the related research on data mining. First of all, it is different from data mining fields based primarily on relational database. We propose a data mining method for native XML database. It can extract some knowledge from native XML database. Secondly, propose a semantic association rule – the rule that is extracted from data mining method. Convert it to the semantic association rule from the proposed procedures so as to make it more legible and easier to users as reference. Finally, propose a database compression using association rules mining. The method compresses the database for reducing the cost of storage. And from the association rules mining, it finds the association among these data. These association rules are further taken as reference for the organizations when making their strategic steps.
32
LIN, MING-HUNG, and 林銘泓. "Exploringthe Distribution Rules of Aggregate Using Data Mining Association Rule." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/00708958833560184595.
Full textAbstract:
碩士萬能科技大學
資訊管理研究所在職專班
104
Aggregate of ready-mixed concrete from the shipping dock to bulk cargo, then vehicle distribution to various ready-mix plant, temporary storage yard. Provided that the transportation process often because there was no effective distribution rules can refer to, lead to a pier laden vehicle waiting distribution caused by congestion. This study by the association rules of data mining methods to retrieve various schedules, content delivery and distribution locations, and thus the formation of the basket, with the relevance of interrelated rules refer to find it. In this study, the use of association rules rule the aggregate distribution is obtained, only that the same timetable and distribution of goods loaded reference rule, if delivery mainland thirds stone, they will delivery six points continent stone; and distribution Hualien sand, it must distribution will Hualien Hualien sixth of stone or stone-thirds. Whereby rules can help dispatchers to quickly make a correct and efficient delivery schedule, another of the study were not included because of the time it is not possible depth information delivery order.
33
Yang, Nai-Hua, and 楊乃樺. "Mining Multidimensional Association Rules for Market Segmentation." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/78f64c.
Full textAbstract:
碩士銘傳大學
資訊管理學系碩士班
95
Today is a customer-oriented market. Enterprises need to give every customer appropriate service. The more precise information can make accurate and profitable strategies. Association rules provide correlations between data items in large numbers of data. The further exploration is to discover relationship between customer’s features and customer purchasing behaviors. This paper proposes a new method to discover mining multidimensional association rule for market segmentation. We use conditional databases to discover multidimensional association rule, do not scan the target database many times and combine cluster method to automatically discretize numerical-type attributes. Our method analyzes CRM data from two different points of view. One is the product combinations according to different customer features; another is the customer features according to purchased products of customers. These two different points of view can provide decision-makers to establish customer profiles, segment market and make strategies more accurately.
34
Jen-Feng, Li. "Mining Association Rules in Time-series Databases." 2005. http://www.cetd.com.tw/ec/thesisdetail.aspx?etdun=U0001-2507200511203800.
Full text
35
Ying-Hsiang, Wen. "Parallel Hardware Architecture for Mining Association Rules." 2006. http://www.cetd.com.tw/ec/thesisdetail.aspx?etdun=U0001-2507200610192300.
Full text
36
Wang, Hsing-Kai, and 王星凱. "An Efficient Distributed Association Rules Mining System." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/75735815815407828838.
Full textAbstract:
碩士淡江大學
資訊管理學系
91
Association rule mining can help the enterprises to capture the consumer behaviors and develop effective marketing strategies. However, the size of transaction database is increasing everyday, how to get timely mining results becomes a serious problem. In this paper, we propose an Effective Distributed Association rule Mining System, EDAMS, to cope with this problem. Unlike other distributed mining systems, a dedicated node is used as data server to collect exchange data among nodes. Thus, the point-to-point broadcasts are avoided and therefore the number of message exchanged is greatly reduced from O(n2) to O(n). Besides, to reduce the total amount of message, the DHP algorithm[2] is used as the basis algorithm to reduce the number of candidate 2-itemsets. According to our experimental results, the EDAMS achieve steadily increasing speedup ration ranging from 100,000 to 700,000 transaction data. Also, the speedup ratio is superior to those in the previous work[7][9]. It clearly demonstrates the effectiveness of our system.
37
Li, Jian-Ming, and 李建明. "Mining Quantitative Association Rules in Disease Databases." Thesis, 2000. http://ndltd.ncl.edu.tw/handle/76433798386921713653.
Full textAbstract:
碩士國立臺灣大學
電機工程學研究所
88
With the computerization of medical information and popularity of medical database, the amount of data grows much more rapidly than ever. There must be numerous known or unknown information hidden behind these data. Traditional statistical approach is not suit for processing such large amount of data. A technique called “Data Mining” is emerging in which the “Association Rules” is the one focusing on the relationship among data items. The technique of mining association rules was first introduced to search the pattern of items that a customer may buy in a supermarket. It can also be extended for mining association rules from a relational database. There are two kinds of attributes in a relational database, one is quantitative and the other is categorical. In this thesis, we introduce a statistical method to finely partition the values of a quantitative attribute into a set of intervals. Different from the previous method which equally partitions the range of an attribute, we suggest a method based on the observation of the data distribution. And we use the mean and standard deviation of each attribute as two parameters of partition. This choice reflects the bias of databases so that it can improve the effectiveness of analysis in highly skewed data. To demonstrate the feasibility of our method, we combine two effective rule-mining algorithms called the DHP algorithm and the Boolean algorithm. With the combination, we can mine association rules from the relational database. Finally, we use this approach on two disease databases. We show the experimental results and compare them with previous methods. The results reveal that our method generated less noises and it was executed easier.
38
Wen, Ying-Hsiang, and 溫英翔. "Parallel Hardware Architecture for Mining Association Rules." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/84140141294459812790.
Full textAbstract:
碩士國立臺灣大學
電機工程學研究所
94
Generally speaking, to implement Apriori-based association rule mining in hardware, one has to load candidate itemsets and a database into the hardware. Since the capacity of the hardware architecture is fixed, if the number of candidate itemsets or the number of items in the database is larger than the hardware capacity, the items are loaded into the hardware separately. The time complexity is in proportion to the number of candidate itemsets multiplied by the number of items in the database. Too many candidate itemsets and a large database would create a performance bottleneck. In this thesis, we propose a HAsh-based and PiPelIned architecture (abbreviated as HAPPI) for hardware-enhanced association rule mining. We apply the pipeline methodology in the HAPPI architecture to compare itemsets with the database and collect useful information for reducing the number of candidate itemsets and items in the database simultaneously. When the database is fed into the hardware, candidate itemsets are compared with the items in the database to find frequent itemsets. At the same time, trimming information is collected from each transaction. In addition, itemsets are generated from transactions and hashed into a hash table. The useful trimming information and the hash table enable us to reduce the number of items in the database and the number of candidate itemsets. Therefore, we can effectively reduce the frequency of loading the database into the hardware. As such, HAPPI solves the bottleneck problem in Apriori-based hardware schemes. We also derive some properties to investigate the performance of this hardware implementation. As shown by the experiment results, HAPPI significantly outperforms the previous hardware approach in terms of execution cycles.
39
Tung, Chien-Hung, and 董建弘. "Incremental XML Association Rules Mining MiningUsing XQuery." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/55107331051607385174.
Full textAbstract:
碩士國立高雄第一科技大學
資訊管理所
94
ABSTRACT XML has already been recognized as a standard for electronic data interchange over the Internet. We believe a large amount of data will be represented and stored in XML format in the near future. Therefore, we think it will be indispensable to develop tools for mining information directly from XML data. Although many well-known data minig methods have been developed, they are almost based on relational database formats. In this paper, we will propose an algorithm, called IXARM (Incremental XML Association Rules Mining), which does not only extract association rules from XML documents, but also offer flexible incremental mining tasks. That is, even when there are many INSERT, DELETE or UPDATE events performed on source XML documents, our algorithm re-computes the modified part only and then combines the previous result to build new association rules in an efficient way. We also have conducted experiment to show the accuracy and performances are all satisfactory for most of the assocation rule mining applications on XML documents.
40
Wan-chuen, Lin, and 林琬純. "Mining Association Rules with Multi-dimensional Constraints." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/65164683786356321966.
Full textAbstract:
碩士國立臺灣大學
資訊管理研究所
91
Association rule mining is an important issue in the area of data mining. Frequent itemset discovery is the key factor in performance of association rule mining. Frequent itemset mining algorithms often generate a very large number of frequent itemsets and rules, which reduce not only the efficiency but also the effectiveness of the mining algorithms since only the subset of the complete frequent itemsets and association rules is of interest to users, and users need additional post-processing to filter through a large number of mined rules to find the useful ones. Constraint-based mining enables users to concentrate on mining itemsets that are interesting to themselves, which improves the efficiency of mining tasks. Previously proposed methods consider that items in transactions are characterized only by single attribute value. In the real world, users may want to keep records of items with respect to more than one attribute and impose constraints on multiple dimensional attributes. In this thesis, we enhance the item representation by associating items with a number of attributes, so-called multi-dimensional items. We have defined and characterized some multi-dimensional constraints. Moreover, we have discussed the properties of those constraints and developed algorithms E-CFG and its generic form, GE-CFG, for mining frequent itemsets with multi-dimensional constraints. The experimental results show that both E-CFG and GE-CFG algorithms outperform the FP-growth+ algorithm for all the cases.
41
Han, Hui Ching, and 韓惠靜. "Mining Association Rules Among Time-series Databases." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/66580094049643941505.
Full textAbstract:
碩士國立臺灣大學
資訊管理研究所
91
While more and more data generated in the form of time-series, there are much more needs to find frequent patterns in time-series data. Time-series data mining becomes more and more popular in recent research areas and has broad applications like analysis of customer purchase patterns, web traversal patterns, etc. Let’s consider the example of stock price fluctuation and trading volume fluctuation patterns of TWSE. There may be some implications that when the stock price went upwards two days before, the stock trading volume may go upwards in following days. The price and trading volume of some leading companies may also affect those changes of other companies in the same industry. It’s interesting for us to find the relationships between stock fluctuation patterns. If we could find out some association rules between stock fluctuation patterns, we can predicate more precisely the trends of stock markets. In this thesis, we propose an algorithm to mine the association rules among time-series data. We view the transaction data describing different attributes or subjects as lines, and then we find association rules among those lines. We’ll introduce a method to find the frequent lines efficiently by constructing the bitmaps of frequent patterns, the method is helpful to reduce the number of database scans. The experimental results show that our proposed algorithm outperforms the Apriori-like approach by several orders of magnitude.
42
Hsi, Lo Yuan, and 羅元禧. "The Association Rules used on Web Mining." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/20237348252767485110.
Full textAbstract:
碩士國立臺北大學
企業管理學系
91
With the trend of increasing web population and data, there is existing a common problem that the pages provided on web are not distinguished with others. All of context provided, web structure and navigation path on web are regulated by the designator. The designator had set up the web only in his opinion, most of them can’t make sure if the context on web could fit user’s real demand. This research would discuss how to analyze the data with Data Mining Method. We had dealt with these data in the first step, and then we would find out the relationship between the data, for instance, association rule. Association rule is one kind of the Data Mining Methods, its primary objective is to find out the association between some specific data items. Because it’s widely accepted by the public, it had been applied to several fields. Using association rule to analyze the registered data makes the result is much fit to general web’s users. Comparing with these three kinds of association rule methods application, the former two methods are applied to web mining in order to find out web association. But the limitation of research method had made those methods application impossible. The second method is to make a solution with long linear method, but it still exists some problems that the conclusions are not in unanimity. The latest method is solve this problem with applying Markov Chain Monte Carlo method to web mining and could include the association between 11 pages in one time. We had used the daily records of NTPU Business Administration Department web to support our research. With the three kinds methods to analyze many navigating information of web users to provide customer-orientation service and information needed. We could sort out these webs base on our conclusion and that would be helpful to set up webs.
43
郭瑞男. "Mining Quantitative Association Rules with Density Constraint." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/13322880493417591245.
Full textAbstract:
碩士國立臺灣師範大學
資訊教育研究所
91
A new approach, called PQAR (Partition-based Quantitative Association Rules mining) algorithm, is proposed in this thesis for mining quantitative association rules. This approach finds out all the frequent interval itemsets that satisfy the minimum relative density requirement based on space partitioning method, and the quantitative association rules are produced from these interval itemsets. When mining frequent interval itemsets, PQAR algorithm considers not only the minimum support as the filtering condition, but also the minimum relative density to prevent finding the intervals in which data distribution is sparse. In addition, based on space partitioning method to find out the largest intervals that meet the threshold requirements, the number of qualified intervals is reduced such that the resulting rules are significant and concise. Furthermore, because the number of times to scan database is reduced possibly in PQAR algorithm, the mining time is shorten considerably than the previous approaches. The experimental results show that, when testing data sets with various supports and relative densities setting, PQAR algorithm obtains results with high accuracy and recall in most cases. Moreover, under the same accuracy condition, PQAR algorithm takes much less time than QAR algorithm.
44
Wu, Chiech-Ming, and 吳界明. "Mining Generalized Association Rules Using Pruning Techniques." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/81678867339436425832.
Full textAbstract:
碩士國立雲林科技大學
電子與資訊工程研究所碩士班
90
The goal of the thesis is to mine generalized association rules using pruning techniques. Given a large transaction database and a hierarchical taxonomy tree of the items, we try to find the association rules between the items at different levels in the taxonomy tree under the assumption that original frequent itemsets and association rules have already been generated beforehand. The primary challenge of designing an efficient mining algorithm is how to make use of the original frequent itemsets and association rules to directly generate new generalized association rules, rather than rescanning the database. In the proposed algorithms GMAR and GMFI, we use join methods and/or pruning techniques to generate new generalized association rules. Through several comprehensive experiments, we find that both algorithms are much better than BASIC and Cumulate algorithms, since they generate fewer candidate itemsets, and furthermore the GMAR algorithm prunes a large amount of irrelevant rules based on the minimum confidence.
45
Shu-Chen, Chang, and 張淑貞. "Mining Multilevel Association Rules in Large Databases." Thesis, 2001. http://ndltd.ncl.edu.tw/handle/62587213281806320843.
Full textAbstract:
碩士國立臺灣科技大學
資訊管理系
89
As a useful business intelligence tool, data mining is gaining its popularity in many real life applications. Association rules is an important data mining technique developed for looking for the relationship among items bought by customers in a supermarket. To increase its applicability, some researchers have worked on mining association rules from relational databases. Mining association rules from relational databases is much more difficult than mining association rules from transactional databases, especially when numerical attributes are involved. This problem is further complicated if one tries to mine rules with concept hierarchy in the domains of the attributes. In this thesis, we study the problem of mining association rules with concept hierarchy in relational databases. For a numerical attribute, we propose an effective clustering algorithm to partition the domain of the attribute such that meaningful association rules can be generated. In the algorithm, we allow the user to give an initial partition on the domain of an attribute. Then, by considering the effect of data skew, we adjust the partitions properly to enhance the chance of generating meaningful rules. For a categorical attribute, we encode the concept hierarchy among values of the attribute, and propose a Boolean- based algorithm for mining association rules with concept hierarchy. We implement the proposed algorithm, test it against different datasets, and study its performance in generating multilevel association rules.
46
Tseng, Pei Hsi, and 曾佩熙. "A Study of Mining Association Rules with." Thesis, 2001. http://ndltd.ncl.edu.tw/handle/61122980878146152093.
Full textAbstract:
碩士臺南師範學院
資訊教育研究所
89
In recently years, as the technology innovates and progresses constantly, computers have been utilized in various walks of life. And, because of the rapid progress on Internet and on World Wide Web, it is more convenient for us to get the information. Consequently, in the case of more and more data, a lot of organizations begin to use database systems to store and manage the data they need. As the rapid growth in the size and number of the database, the technology of discovering useful knowledge hidden in the large database has become an important research topic. Data mining is the important task of knowledge discovery in databases. The mining of association rule has become one of the important data mining technology. Association rules can be used to express relationships between items of data. Mining association rules is to analyze the data in a database to discover interesting rules. However, existing algorithms require a record in the database contain all the data items in a rule. Such as the rule {A}→{B,C}, a transaction which contains A, B, and C will be counted. This requirement makes it difficult to discover certain useful rules in some applications. For example, a patient may not show all symptoms of a disease so that he does not keep the right time to cure. To solve the problem, a method which can support mining association rules with composite items was proposed. However, in order to find all interesting composite items, it uses a more complicated method to prune. There, in this thesis, we will propose a new mining strategy, which is called SimplePrune approach to reduce the overhead of pruning. In addition, we supports other Boolean operations(such as AND and NOT)and special types of binary operations(such as XOR and XNOR). Via our proposed method, various types of association rules between composite items can be discovered.
47
Qui, Ding-Ying, and 邱鼎穎. "Mining Weighted Association Rules from Large Database." Thesis, 2001. http://ndltd.ncl.edu.tw/handle/69625187682793463209.
Full textAbstract:
碩士輔仁大學
資訊工程學系
89
Mining association rules is to find associations among items from large transaction databases. Weighted association rule is also to describe the associations among items, and the importance of each item is considered. Hence, weighted association rules can provide more information than that of association rules. There are many researchers that have proposed their algorithms for mining association rules. Apriori algorithm needs to scan database many times and cost much search time for mining association rules, which is very inefficient. Many other algorithms improved the efficiency of Apriori algorithm, but a lot of memory space need to be taken. There are few approached proposed for mining weighted association rules. The previous approaches also take a lot of time to scan database and search for the needed information. In this thesis, we propose two new algorithms --- delay transaction disassemble (DTD) and Weighted Daley Transaction Disassembled (WDTD) for mining association rules and weighted association rules, respectively, which is very efficient and need not take much memory space. We observe that most of time spent for scanning the database is to disassemble each transaction in the database. The main idea of DTD and WDTD algorithms are that the transaction is not disassembled until it needs to be used. For mining weighted association rules, we also propose an AVL-index tree to store the transactions and the transaction overlap technique to further reduce the execution time and the memory space. The experimental results show that our algorithms outperform other algorithms for mining association rules and weighted association rules.
48
Ling, Chun-Ching, and 凌俊青. "Mining Quantitative Association Rules in Bag Databases." Thesis, 1999. http://ndltd.ncl.edu.tw/handle/64760497556217622481.
Full textAbstract:
碩士國立中央大學
資訊管理研究所
87
The problem of mining association rules is to find the associations between items in a large database of sales transactions. Although there are a lot of previous researches on this area, a common problem occurred is that the rule only indicates if two items are related but as to in what quantities and in what combinations are missing. Without this information it is impossible to design a competitive combination of sales items since we didn''t know how many units of items should be included. Therefore, if the quantities of items can be included in association rules, it will be helpful for managers to make the marketing decisions. In this paper, we introduce a new algorithm for mining association rules including the quantities of items. Then, we extend the rules so that the quantities of items can be expressed as user-defined intervals or fuzzy terms.
49
chien, I.-pei, and 錢依佩. "An Efficient Algorithm for Mining Association Rules." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/94139240649049674349.
Full textAbstract:
碩士南台科技大學
資訊管理系
91
Due to the improvement of information technologies and popularization of computers, collecting information becomes easier, rapider and more convenient than before. As the time goes by, database cumulates huge and hiding information. Therefore, how to correctly uncover and efficiently mining from those hiding information becomes a very important issue. Hence the technology of data mining becomes one of the solutions. In the technologies of data mining, association rules mining is one of the most popular technology to be used. Association rule mining explores the approaches to extract the frequent itemsets from large database. Further, derives the knowledge behind implicitly. The Apriori algorithm is one of the most frequently used algorithms. Although the Apriori algorithm can successful derive the association rules from database, the Apriori algorithm has two major defects: First, the Apriori algorithm will produce large amounts of candidate itemsets during extracting the frequent itemsets from large database. Second, frequently scanning whole database lead to inefficient performance. Many researches try to improve the performance of the Apriori algorithm, but still not escape from the frame of the Apriori algorithm and lead to a little improvement of the performance. In this paper we propose QDT and ICI which escape the frame of Apriori algorithm, and it only needs to scan whole database once during extracting the frequent itemsets from large database. Therefore, the QDT and ICI algorithm can efficiently reduce the I/O time, and rapidly extract during extracting the frequent itemsets from large database, and make data mining more efficient than before.
50
Lai, Wei Lin, and 賴韋霖. "Improving Apriori Algorithms for Mining Association Rules." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/42956763049200327986.
Full textAbstract:
碩士南台科技大學
資訊管理系
97
With the development of information technology, enterprises have a lot of way to get information and can use this technology store about a lot of enterprise’s transaction or record in data base. How to find the useful information in database has become the subject which the enterprises pay attention. Association rules technology is generally in data mining. Based on the Internet Technology development and the globalization of business, the transaction database of enterprise is constantly changing all the time, and in order to keep the accuracy of exploring result in dynamic database, the traditional explore method in order to keep the information accuracy so it unavoidable must to exploring information again constantly; Because generated too many redundant candidate itemsets so it causes too many times to scan the database; Is need to scan the redundant transaction data because there is not recognize this items belong to which transaction. In order to preserve the accuracy when mining the dynamic database, we need repeatedly scan database. This is above the traditional Apriori algorithm to mining association rules of the weakness in the dynamic database. This research is based on Apriori Algorithm to improve its process. This paper proposed two improve algorithms. One of VS_Apriori(Vertial Scan Apriori) algorithm is to transform database from horizontal to vertical. This can be avoided scan redundant of Transaction data.Any item count just need to scan two transactions in data base so as to increase mining efficiency. In addition,SVS_Apriori algorithm (Sort & Vertial Scan Apriori)that is improved from Apriori generate candidate itemsets process. First, itemsets sort by ascending so that can avoid generate too many candidate itemset and can increase mining efficiency again. And propose appropriate methods to update these two algorithms so as to these algorithms can use in dynamic database in real-time and correctly, to fit in with the business needs and provide immediate and accurate to the important decision-making.