Kusakcı, Ali OsmanYazgan, Pınar2016-11-302018-08-052016-11-302018-08-052016Yazgan, Pınar. (2016). Association rules and market baset analysis : a case study in retail sector. (Yayımlanmamış yüksek lisans tezi). İstanbul Ticaret Üniversitesi, Fen Bilimleri Enstitüsü, Endüstri Mühendisliği, İstanbulhttps://hdl.handle.net/11467/1997http://library.ticaret.edu.tr/e-kaynak/tez/68737.pdfTez (Yüksek Lisans) -- İstanbul Ticaret Üniversitesi -- Kaynakça var.Hızla gelişen teknoloji sayesinde marketler ve işletmeler verilerini kolayca saklayabilmektedirler. Gerçekleştirilen her işlem depolanarak veri setlerini oluşturmaktadır. Gittikçe büyüyen bu veri setlerinden yararlı bilgiler elde edilme si gerekmektedir. İşte bu aşamada veri madenciliği devreye girmektedir. Bu çalışmada, öncelikle veri madenciliğinin temelleri, aşamaları, kullanım alanları ve temel algoritma çeşitlerinden bahsedilmiştir. Daha sonra Veri Madenciliği modellerinden olan “Birliktelik Kuralları” algoritmaları üzerinde durulmuş, bu algoritmalar arasında bir değerlendirme yapılarak Apriori algoritması tercih edilmiştir. Son bölümde Türkiye’deki bir perakende satış mağazasının verileri ile Apriori algoritması kullanılarak Birliktelik Kuralı Analizi uygulanmış ve ürünler arasındaki ilişkiler ortaya çıkarılmıştır.Thanks to the rapidly developing technology, companies and businesses can easily store their data. Storage of each transaction performed forms data sets. Useful informations has to be obtained from the steadily growing data sets. At this stage, data mining is of paramount importance. Firstly, in this study, the basics of data mining, its stages, application areas and types of basic algorithms are discussed. Secondly, a comprehensive review of "Association Rules" algorithms, as one of the main tools of data mining, is presented. Considering the strenghts and weaknesses of the presented algorithms, Apriori algorithm is preferred for application. Lastly, association rule analysis is applied using Apriori Algorithm on data of one of the retail chains in Turkey and relations between products are revealed. Furthermore, the implicaitons of the analysis are discussed in detail.CONTENTS, i -- SUMMARY, ii -- ABSTRACT, iii -- THANKS, iv -- INDEX OF FIGURES, v -- INDEX OF TABLES, vi -- INDEX OF SYMBOLS AND ABBREVIATIONS, vii -- 1. INTRODUCTION, 1 -- 1.1 The Subject And Scope, 1 -- 1.2 Purpose And Importance, 2 -- 2. LITERATURE REVIEW, 4 -- 2.1 Frequent Itemset Mining, 5 -- 2.1.1 Algorithms for mining from horizontal layout database, 6 -- 2.1.1.1 Apriori algorithm, 7 -- 2.1.1.2 Direct hashing and pruning (DHP) algorithm, 9 -- 2.1.1.3 Partitioning algorithm, 9 -- 2.1.1.4 Dynamic itemset counting algorithm (DIC), 10 -- 2.1.1.5 Sampling algorithm, 10 -- 2.1.1.6 Continuous association rule mining algorithm (CARMA), 10 -- 2.1.1.7 Split and merge algorithm (SAM), 11 -- 2.1.1.8 PRICES algorithm, 11 -- 2.1.2 Algorithms based on vertical layout database, 11 -- 2.1.2.1 Equivalence class transformation algorithm (ECLAT), 11 -- 2.1.3 Algorithms for mining from projected layout based database, 12 -- 2.1.3.1 FP_growth algorithm, 12 -- 2.1.3.2 H_mine algorithm, 13 -- 2.2 Sequential Pattern Mining, 15 -- 2.2.1 Apriori based approaches (the candidate generation-and-test approach), 16 -- 2.2.1.1 Generalized sequential patterns algorithm (GSP), 16 -- 2.2.1.2 Sequential pattern discovery using equivalent classes algorithm (SPADE), 16 -- 2.2.2 Pattern-growth-based approaches, 16 -- 2.2.2.1 Frequent pattern-projected sequential pattern mining (FREESPAN), 17 -- 2.2.2.2 prefix-projected sequential patterns mining (PrefixSpan), 17 -- 2.3 Structured Pattern Mining, 19 -- 2.3.2 Frequent subgraph discovery algorithm (FSG), 19 -- 2.3.3 Graph-based substructure pattern mining algorithm (GSPAN), 18 -- 2.3.1 SUBDUE algorithm, 19 -- 2.3.4 Inductive logic programming algorithm (WARMR), 19 -- 3. DATA MINING, 21 -- 3.1 What Is Data Mining?, 21 -- 3.2 History Of Data Mining, 21 -- 3.3 Data Mining Process, 23 -- 3.3.1 Defining the problem, 24 -- 3.3.2 Preparation of data, 24 -- 3.3.2.1 Collection, 25 -- 3.3.2.2 Assessment, 25 -- 3.3.2.3 Consolidation and cleaning, 25 -- 3.3.2.4 Selection, 25 -- 3.3.2.5 Transformation, 25 -- 3.3.3 Establishment and evaluation of the model, 25 -- 3.3.4 Using the model, 26 -- 3.3.5 Monitoring model, 26 -- 3.4 Data Mining Areas, 26 -- 3.5 The Problems in Data Mining (Savas et al., 2012), 28 -- 3.6 Data Mining Algorithms, 29 -- 3.6.1 Predictive models, 29 -- 3.6.1.1 Regression, 29 -- 3.6.1.2 Classification, 30 -- 3.6.1.2.1 Decision trees, 30 -- 3.6.1.2.2 Naive bayes classification algorithm, 31 -- 3.6.1.2.3 Time series algorithm, 32 -- 3.6.1.2.4 Genetic algorithms, 33 -- 3.6.1.2.5 Artificial neural networks, 33 -- 3.6.2 Descriptive models, 34 -- 3.6.2.1 Clustering method, 34 -- 4. ASSOCIATION RULE MINING, 36 -- 4.1 Commonly Used Terms, 37 -- 4.2 Creation of Association Rules, 39 -- 4.3 Successfull Points Of Market Basket Analysis, 40 -- 4.4 Failed Points Of Market Basket Analysis, 40 -- 4.5 Types Of Association Rules, 41 -- 4.5.1 Types of values handled, 41 -- 4.5.2 Levels of abstraction involved, 41 -- 4.5.3 Dimensions of data involved, 42 iii -- 5. A CASE STUDY OF MBA ON A SUPERMARKET CHAIN, 43 -- 5.1 Dataset, 43 -- 5.2 Methods And Algorithms Used, 44 -- 5.2.1 Apriori algorithm, 44 -- 5.2.2 Apriori algorithm solution steps, 46 -- 5.2.3 Application areas of apriori algorithm, 50 -- 5.2.4 Program, 52 -- 7. CONCLUSIONS AND RECOMMENDATIONS, 58 -- REFERENCES, 62 -- CURRICULUM VITAE (CV), 69eninfo:eu-repo/semantics/openAccessVeri madenciliğiPazarlama_Veri işlemİş_Veri işlemData miningMarketing_Data processingBusiness_Data processingQA 76.9.D343/Y39Master Thesis184442476