Re-mining item associations: Methodology and a case study in apparel retailing

Demiriz, Ayhan; Ertek, Gürdal; Atan, Tankut; Kula, Ufuk

doi:10.1016/j.dss.2011.08.004

Cited by 21 publications

(9 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Liu and Yin [34] presented an efficient data re-mining technique for updating previously discovered association rules in the light of threshold changes. Demiriz et al [35] proposed a new approach to mine price-, time-, and domain-related attributes through re-mining of association mining results. The underlying factors behind positive and negative relationships were characterized and described in a second data mining stage.…”

Section: Related Workmentioning

confidence: 99%

DTreeSim: A new approach to compute decision tree similarity using re-mining

Bakırlı¹,

Birant²

2017

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

A number of recent studies have used a decision tree approach as a data mining technique; some of them needed to evaluate the similarity of decision trees to compare the knowledge reflected in different trees or datasets. There have been multiple perspectives and multiple calculation techniques to measure the similarity of two decision trees, such as using a simple formula or an entropy measure. The main objective of this study is to compute the similarity of decision trees using data mining techniques. This study proposes DTreeSim, a new approach that applies multiple data mining techniques (classification, sequential pattern mining, and k-nearest neighbors) sequentially to identify similarities among decision trees. After the construction of decision trees from different data marts using a classification algorithm, sequential pattern mining was applied to the decision trees to obtain rules, and then the k-nearest neighbor algorithm was performed on these rules to compute similarities using two novel measures: general similarity and pieced similarity.Our experimental studies compared the results of these novel similarity measures and also compared our approach with existing approaches. Our comparisons indicate that our proposed approach performs better than existing approaches, because it takes into account the values of the branches in the trees through sequential pattern mining.

show abstract

Section: Related Workmentioning

confidence: 99%

DTreeSim: A new approach to compute decision tree similarity using re-mining

Bakırlı¹,

Birant²

2017

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

show abstract

“…The idea was that, all item sets cannot be expected to behave similarly and hence it would be more appropriate to evaluate the behavior of different item sets using different minimum support values. In [7][8][9][10][11][12][13][14][15][16][17][18], methods were published for eliminating the less interesting rules which is one of the most challenging task in mining negative association rules. In [19,20], methods were published to mine negative association rule by minimizing the number of scans of the dataset.…”

Section: Related Workmentioning

confidence: 99%

EO-ARM: An efficient and optimized k-map based positive-negative association rule mining technique

Ravi

Khare

2014

2014 International Conference on Circuits, Power and Computing Technologies [ICCPCT-2014]

View full text Add to dashboard Cite

Association Rule Mining is a Data Mining technique which extracts association rules from the given dataset. A good number of research work has been reported in Association Rule Mining algorithms which discovers positive association rules. Amongst them, only a few algorithms have focused on Association Rule Mining algorithms which discovers negative association rules too. Amongst the negative Association Rule Mining algorithms, most of them scans the dataset more than once to identify the frequent item sets and also doesn't guarantee that all the extracted rules are interesting. In order to overcome the above said challenges, EO-ARM, an Efficient and Optimized Positive-Negative Association Rule Mining algorithm has been proposed in this paper. EO-ARM produces both positive as well as negative association rules. It scans the dataset only once (irrespective of the size of dataset) to identify frequent item sets using a two dimensional matrix thereby increasing the efficiency. The two dimensional matrix is conceptually similar to k-map. It also optimizes the association rules by introducing a contingency matrix based correlation measure which prunes less interesting rules thereby overcoming the existing limitations. EO-ARM has been implemented using a binary transaction dataset. Several experiments were performed and an optimal support and confidence threshold has been identified for the given dataset. These optimized support and confidence threshold have been used to find the frequent item sets and generating rules from the dataset. Experimental results also proved that EO-ARM is more efficient in terms of execution time than the standard Apriori algorithm and more optimized in terms of no. of rules generated with the pruning done with the projected correlation measure.

show abstract

“…A well-known example of such a system is one employed by the online retailer Amazon and recommendations are exposed to users under the header "Customers Who Bought This Item Also Bought". Offline retailers can also employ association mining for developing rules and policies that can increase sales and profits [12]- [14] such as decisions on shelf layout, bundles or last minute offerings near the cashiers can be made based on items which are bought together frequently, such as peanut butter and grape jelly [15], can be found as neighboring items on the shelves because they complement each other.…”

Section: Introductionmentioning

confidence: 99%

Profit estimation error analysis in recommender systems based on association rules

Ertek

Yee

et al. 2015

2015 IEEE International Conference on Big Data (Big Data)

View full text Add to dashboard Cite

It is a challenge to estimate expected benefits from recommender systems based on association rule mining. This paper aims to address this challenge and presents a study of buying preferences of a sample of retail customers. It reveals a monotonic, non-linear relationship between the expected profits (as a function of information loss) and minimum support threshold levels, when considering transactions for a recommender system based on association rules. This finding is significant for recommender systems that utilize potential profits as a decisionmaking criterion.

show abstract

Re-mining item associations: Methodology and a case study in apparel retailing

Cited by 21 publications

References 15 publications

DTreeSim: A new approach to compute decision tree similarity using re-mining

DTreeSim: A new approach to compute decision tree similarity using re-mining

EO-ARM: An efficient and optimized k-map based positive-negative association rule mining technique

Profit estimation error analysis in recommender systems based on association rules

Contact Info

Product

Resources

About