scholarly journals Penerapan Association Rule Mining Berbasis Algoritma Frequent Pattern Growth untuk Rekomendasi Penjualan

2020 ◽  
Vol 7 (2) ◽  
pp. 135-148
Author(s):  
Didi Supriyadi
Keyword(s):  
Data Mining ◽  
Association Rule ◽  
Association Rule Mining ◽  
Frequent Itemset ◽  
Frequent Pattern ◽  
Rule Mining ◽  
Minimum Support ◽  
Pattern Growth ◽  
Mining Association Rule

Tingkat persaingan dan kompleksitas permasalahan penjualan pada perusahaan retail, menuntut setiap perusahaan retail untuk mampu berkompetisi dengan perusahaan lain. Salah satu yang dapat dilakukan adalah melalui pengambilan keputusan terkait penjualan yang lebih tepat dan efektif. Besarnya data transaksinonal penjualan perusahaan retail dapat dilakukan ekstraksi informasi yang bermanfaat. Metode yang dapat digunakan untuk menggali informasi adalah melalui penerapan association rule mining. Association Rule Mining merupakan suatu metode data mining yang berfokus pada pola transaksi dengan cara mengekstraksi asosiasi atau hubungan suatu kejadian. Keranjang belanja yang terdapat pada perusahaan retail yang terkomputerisasi merupakan cara terbaik untuk memberikan dukungan rekomendasi keputusan secara ilmiah dengan cara menentukan hubungan antara barang yang dibeli secara bersamaan dalam setiap transaksi. Algoritma FP-growth digunakan untuk menentukan himpunan dataset yang paling sering muncul (frequent itemset) pada sekeompok data. Penelitian ini menghasilkan nilai minimum support 0,1% dan nilai minimum confidence 60% jumlah rule yang dihasilkan berjumlah 116457, nilai minimum confidence 70% jumlah rule yang dihasilkan berjumlah 84086, dan nilai minimum confidence 80% jumlah rule yang dihasilkan berjumlah 48623 dari data yang diolah sebanyak 22191. Hasil rule ini dapat digunakan untuk strategi pemasaran produk. Nilai minimum support 0,1% dimana semakin besar nilai minimum confidence maka menghasilkan rule yang semakin sedikit.

scholarly journals Analisis Tingkat Pembelian Konsumen dengan Algoritma Apriori

2020 ◽  
Vol 2 (02) ◽  
Author(s):  
Ismasari Ismasari ◽  
Maulida Ramadhan ◽  
Wahyu Hadikristanto
Keyword(s):  
Data Mining ◽  
Association Rule ◽  
Association Rule Mining ◽  
Market Basket Analysis ◽  
Rule Mining ◽  
Market Basket ◽  
Minimum Support ◽  
Mining Association Rule

Saat ini data mining telah diimplementasikan ke berbagai bidang salah satu diantaranya adalah pada bidang bisnis atau perdagangan yang dapat membantu para pebisnis dalam kebijakan pengambilan keputusan terhadap apa yang berhubungan dengan persediaan barang. Misalnya pentingnya sistem persediaan barang di suatu Toko dan jenis barang apa yang menjadi prioritas utama yang harus di stok untuk mengantisipasi kekosongan barang. Karena minimnya stok barang dapat berpengaruh pada pelayanan konsumen dan pendapatan Toko. Metode yang sering digunakan untuk menganalisa pola pembelian pelanggan adalah metode asosiasi atau association rule mining. Association rule mining adalah suatu metode untuk mencari pola hubungan antar satu atau lebih itemset yang ada dalam suatu dataset. Algoritma yang paling popular dalam mencari pola hubungan item set adalah algoritma apriori atau sering disebut dengan market basket analysis. Proses yang dilakukan dalam penelitian ini menggunakan tools Rapid Miner untuk mengolah data dengan algoritma apriori, dari pengujian yang dilakukan dengan parameter yang telah ditentukan yaitu minimum support 70% dan minimum confidence 80% menghasilkan 4 aturan asosiasi dengan nilai confidance 100% yaitu kombinasi item aqua 600ml-fulloblasto caramel cruncy chocolat - yupi 500 semua rasa - beng beng 25g. Dengan pencarian pola menggunakan algoritma apriori ini diharapkan informasi yang dihasilkan dapat meningkatakan strategi penjualan selanjutnya    

scholarly journals Minimum Threshold Determination Method based on Dataset Characteristics in Association Rule Mining

2021 ◽  
Author(s):  
Erna Hikmawati ◽  
Nur Ulfa Maulidevi ◽  
Kridanto Surendro
Keyword(s):  
Association Rule ◽  
Association Rule Mining ◽  
Threshold Value ◽  
Extraction Process ◽  
Frequent Itemset ◽  
Rule Mining ◽  
Minimum Threshold ◽  
Minimum Support ◽  
Associative Behavior

Abstract The process of extracting data to obtain useful information is known as data mining. Furthermore, one of the promising and widely used techniques for this extraction process is association rule mining. This technique is used to identify interesting relationships between sets of items in a dataset and predict associative behavior for new data. The first step in association rule mining is the determination of the frequent item set that will be involved in the rule formation process. In this step, a threshold is used to eliminate items excluded in the frequent itemset which is also known as the minimum support. Furthermore, the threshold provides an important role in determining the number of rules generated. However, setting the wrong threshold leads to the failure of the association rule mining to obtain rules. Currently, the minimum support value is determined by the user. This leads to a challenge that becomes worse for a user that is ignorant of the dataset characteristics. In this study, a method was proposed to determine the minimum support value based on the characteristics of the dataset. Furthermore, this required certain criteria to be used as thresholds which led to more adaptive rules according to the needs of the user. The results of this study showed that 6 from 8 datasets, obtained a rule with lift ratio > 1 using the minimum threshold value that was determined through this method.

Comparative Analysis of Association Rule Mining Algorithms in Market Basket Analysis Using Transactional Data

2020 ◽  
Vol 27 (1) ◽  
Author(s):  
AA Izang ◽  
SO Kuyoro ◽  
OD Alao ◽  
RU Okoro ◽  
OA Adesegun
Keyword(s):  
Data Mining ◽  
Association Rule ◽  
Association Rule Mining ◽  
Market Basket Analysis ◽  
Rule Mining ◽  
Business Decisions ◽  
Market Basket ◽  
Minimum Support ◽  
Support Threshold ◽  
Transactional Data

Association rule mining (ARM) is an aspect of data mining that has revolutionized the area of predictive modelling paving way for data mining technique to become the recommended method for business owners to evaluate organizational performance. Market basket analysis (MBA), a useful modeling technique in data mining, is often used to analyze customer buying pattern. Choosing the right ARM algorithm to use in MBA is somewhat difficult, as most algorithms performance is determined by characteristics such as amount of data used, application domain, time variation, and customer’s preferences. Hence this study examines four ARM algorithm used in MBA systems for improved business Decisions. One million, one hundered and twele thousand (1,112,000) transactional data were extracted from Babcock University Superstore. The dataset was induced with Frequent Pattern Growth, Apiori, Association Outliers and Supervised Association Rule ARM algorithms. The outputs were compared using minimum support threshold, confidence level and execution time as metrics. The result showed that The FP Growth has minimum support threshold of 0.011 and confidence level of 0.013, Apriori 0.019 and 0.022, Association outliers 0.026 and 0.294 while Supervised Association Rule has 0.032 and 0.212 respectively. The FP Growth and Apirori ARM algorithms performed better than Association Outliers and Supervised Association Rule when the minimum support and confidence threshold were both set to 0.1. The study concluded by recommending a hybrid ARM algorithm to be used for building MBA Applications. The outcome of this study when adopted by business ventures will lead to improved business decisions thereby helping to achieve customer retention. Keywords: Association rule mining, Business ventures, Data mining, Market basket analysis, Transactional data.

scholarly journals An Enhanced Approach to Mine Maximal Frequent Itemset using Maximal Frequent Itemset Prima Algorithm (MFIPA)

2019 ◽  
Vol 8 (S2) ◽  
pp. 9-12
Author(s):  
R. Smeeta Mary ◽  
K. Perumal
Keyword(s):  
Data Mining ◽  
Association Rule ◽  
Association Rule Mining ◽  
Frequent Itemsets ◽  
Frequent Itemset ◽  
Decision Makers ◽  
New Method ◽  
Rule Mining

In data mining finding out the frequent itemsets is one of the very essential topics. Data mining helps in identifying the best knowledge for different decision makers. Frequent itemset generation is the precondition and most time-consuming method for association rule mining. In this paper we suggest a new algorithm for frequent itemset detection that works with datasets in distributed manner. The proposed algorithm brings in a new method to find frequent itemset not including the necessitate to create candidate itemsets. The proposed approach could be implemented using horizontal representation for transaction datasets and allocating prime value. It explores all the frequent itemset that is present in the input and according to the support the maximum frequent itemset is identified. It was applied on different transactions database and compared with well-known algorithms: FP-Growth and Parallel Apriori with different support levels. The try out showed that the proposed algorithm attain major time improvement over both algorithms.

scholarly journals Analisa Pola Frekuensi Keranjang Belanja Dengan Dengan Perbandingan Algoritma Fp-Growth (Frequent Pattern Growth) dan Eclat pada minimarket

2020 ◽  
Vol 7 (2) ◽  
pp. 364-373
Author(s):  
Krisna Nata Wijaya
Keyword(s):  
Data Mining ◽  
Association Rule ◽  
Frequent Pattern ◽  
Minimum Support ◽  
Pattern Growth

Dalam kegiatan transaksi jual beli di minimarket ataupun toko pemilik harus mengerti apa yang diinginkan komsumen dalam memberikan kenyaman berbelanja, terutama kemudahan dalam pemilihan barang yang disesuaikan dengan tata letak atau penempatan barang. Dengan menerapkan association rule pada data transaksi akan memudahkan pemilik dalam mengelolah informasi penjualan dan mencari itemset. Oleh karena itu, penelitian ini Melakukan analisis pola data transaksi penjualan dengan menerapkan metode asosiasi pada data mining. Selanjutnya dengan melakukan perbandingan algoritma Fp-Growth dan Eclat dengan minimum support dan confidence sebesar 0.01% untuk menentukan jumlah aturan yang terbentuk sebagai bahan pengambil keputusan yang ditunjukan untuk frekuensi keranjang belanja.

scholarly journals PENERAPAN METODE ASSOCIATION RULE MENGGUNAKAN ALGORITMA APRIORI PADA SIMULASI PREDIKSI HUJAN WILAYAH KOTA BANDUNG

2016 ◽  
Vol 2 (3) ◽  
Author(s):  
Mohamad Fauzy ◽  
Kemas Rahmat Saleh W ◽  
Ibnu Asror
Keyword(s):  
Data Mining ◽  
Data Processing ◽  
Association Rules ◽  
Association Rule ◽  
Association Rule Mining ◽  
Weather Forecast ◽  
Weather Data ◽  
Rule Mining ◽  
Similarity Relationships ◽  
Mining Association Rule

[Id] Prakiraan cuaca saat ini telah menjadi satu hal yang dibutuhkan bagi banyak orang di dunia. Dalam memprediksi hujan pengolahan data cuaca merupakan hal yang penting. Namun permasalahannya, data cuaca yang semakin hari semakin bertambah menyebabkan penumpukan data sehingga pengolahan data tersebut perlu penanganan lebih lanjut. Oleh karena itu pemanfaatan data mining digunakan untuk menyelesaikan masalah ini. Association rule mining adalah salah satu metode data mining yang dapat mengidentifikasi hubungan kesamaan antar item. Penelitian ini dilakukan dengan tiga tahapan utama yaitu : 1) melakukan analisa pola frekuensi tinggi menggunakan algortima apriori; 2) pembentukan aturan asosiasi (association rule); 3) uji kekuatan rule yang terbentuk dengan menghitung lift ratio pada masing-masing rule. Dataset yang digunakan adalah data klimatologi yang diambil dari BMKG stasiun geofisika kelas 1 Bandung. Hasil akhir dari Penelitian ini berupa aturan-aturan asosiasi (association rules) dimana aturan-aturan ini dapat dijadikan sebagai acuan dalam memprediksi cuaca hujan atau tidak hujan untuk satu hari kedepan. Kata kunci : Data mining, association rule, apriori, prediksi hujan [En] Weather forecast today has become a necessary thing for many people in the world. In predicting rain weather data processing is essential. But the problem, weather data that is increasingly growing cause the accumulation of data so that the data processing needs further treatment. Therefore, the use of data mining is used to solve this problem. Association rule mining is one of data mining methods that can identify similarity relationships between items. This research is performed by three main stages, namely: 1) to analyze high frequency patterns using algorithms priori; 2) the establishment of an association rule (association rule); 3) test the strength of the rule which is formed by calculating the ratio elevator on each rule. The dataset used is the climatological data taken from BMKG station 1st class geophysical Bandung. The end result of this research in the form of rules of association (association rules) in which these rules can be used as a reference in predicting the weather is rain or not rain for the next day. Keywords : data mining, association rule, apriori, rain forecast

scholarly journals Algoritma Apriori untuk Pencarian Frequent itemset dalam Association Rule Mining

2019 ◽  
Vol 7 (2) ◽  
pp. 143-152
Author(s):  
Lusa Indah Prahartiwi ◽  
Wulan Dari
Keyword(s):  
Data Mining ◽  
Association Rule ◽  
Association Rule Mining ◽  
Frequent Itemset ◽  
Data Availability ◽  
Department Stores ◽  
Frozen Foods ◽  
Rule Mining ◽  
Snack Foods ◽  
Transactional Data

Abstract   Over decades, retail chains and department stores have been selling their products without using the transactional data generated by their sales as a source of knowledge. Abundant data availability, the need for information (or knowledge) as a support for decision making to create business solutions, and infrastructure support in the field of information technology are the embryos of the birth of data mining technology. Association rule mining is a data mining method used to extract useful patterns between data items. In this research, the Apriori algorithm was applied to find frequent itemset in association rule mining. Data processing using Tanagra tools. The dataset used was the Supermarket dataset consisting of 12 attributes and 108.131 transaction. The experimental results obtained by association rules or rules from the combination of item-sets beer wine spirit-frozen foods and snack foods as a Frequent itemset with a support value of 15.489% and a confidence value of 83.719%. Lift ratio value obtained was 2.47766 which means that there were some benefits from the association rule or rules.   Keywords: Apriori, Association Rule Mining.   Abstrak   Selama beberapa dekade rantai ritel dan department store telah menjual produk mereka tanpa menggunakan data transaksional yang dihasilkan oleh penjualan mereka sebagai sumber pengetahuan. Ketersediaan data yang melimpah, kebutuhan akan informasi (atau pengetahuan) sebagai pendukung pengambilan keputusan untuk membuat solusi bisnis, dan dukungan infrastruktur di bidang teknologi informasi merupakan cikal-bakal dari lahirnya teknologi data mining. Data mining menemukan pola yang menarik dari database seperti association rule, correlations, sequences, classifier dan masih banyak lagi yang mana association rule adalah salah satu masalah yang paling popular. Association rule mining merupakan metode data mining yang digunakan untuk mengekstrasi pola yang bermanfaat di antara data barang. Pada penelitian ini diterapkan algoritma Apriori untuk pencarian frequent itemset dalam association rule mining. Pengolahan data menggunakan tools Tanagra. Dataset yang digunakan adalah dataset Supermarket yang terdiri dari 12 atribut dan 108.131 transaksi. Hasil eksperimen diperoleh aturan asosiasi atau rules dari kombinasi itemsets beer wine spirit-frozen foods dan snack foods sebagai Frequent itemset dengan nilai support sebesar 15,489% dan nilai confidence sebesar 83,719%. Nilai Lift ratio yang diperoleh sebesar 2,47766 yang artinya terdapat manfaat dari aturan asosiasi atau rules tersebut.   Kata kunci: Apriori, Association rule mining  

scholarly journals Present State-of-The-ART of Dynamic Association Rule Mining Algorithms

2019 ◽  
Vol 9 (1) ◽  
pp. 309-316
Keyword(s):  
Data Mining ◽  
Association Rules ◽  
Association Rule ◽  
Association Rule Mining ◽  
Rule Mining ◽  
Healthcare Data ◽  
Data Mining Approach ◽  
Mining Community ◽  
Pattern Growth ◽  
Mining Algorithms

Association Rule Mining (ARM) is a data mining approach for discovering rules that reveal latent associations among persisted entity sets. ARM has many significant applications in the real world such as finding interesting incidents, analyzing stock market data and discovering hidden relationships in healthcare data to mention few. Many algorithms that are efficient to mine association rules are found in the existing literature, apriori-based and Pattern-Growth. Comprehensive understanding of them helps data mining community and its stakeholders to make expert decisions. Dynamic update of association rules that have been discovered already is very challenging due to the fact that the changes are arbitrary and heterogeneous in the kind of operations. When new instances are added to existing dataset that has been subjected to ARM, only those instances are to be used in order to go for incremental mining of rules instead of considering the whole dataset again. Recently some algorithms were developed by researchers especially to achieve incremental ARM. They are broadly grouped into Apriori-based and Pattern-Growth. This paper provides review of Apriori-based and Pattern-Growth techniques that support incremental ARM.

scholarly journals Minimum threshold determination method based on dataset characteristics in association rule mining

2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Erna Hikmawati ◽  
Nur Ulfa Maulidevi ◽  
Kridanto Surendro
Keyword(s):  
Association Rule ◽  
Association Rule Mining ◽  
Threshold Value ◽  
Frequent Itemset ◽  
Rule Mining ◽  
Minimum Threshold ◽  
Minimum Support ◽  
Associative Behavior ◽  
Average Utility ◽  
Adaptive Support

AbstractAssociation rule mining is a technique that is widely used in data mining. This technique is used to identify interesting relationships between sets of items in a dataset and predict associative behavior for new data. Before the rule is formed, it must be determined in advance which items will be involved or called the frequent itemset. In this step, a threshold is used to eliminate items excluded in the frequent itemset which is also known as the minimum support. Furthermore, the threshold provides an important role in determining the number of rules generated. However, setting the wrong threshold leads to the failure of the association rule mining to obtain rules. Currently, user determines the minimum support value randomly. This leads to a challenge that becomes worse for a user that is ignorant of the dataset characteristics. It causes a lot of memory and time consumption. This is because the rule formation process is repeated until it finds the desired number of rules. The value of minimum support in the adaptive support model is determined based on the average and total number of items in each transaction, as well as their support values. Furthermore, the proposed method also uses certain criteria as thresholds, therefore, the resulting rules are in accordance with user needs. The minimum support value in the proposed method is obtained from the average utility value divided by the total existing transactions. Experiments were carried out on 8 specific datasets to determine the association rules using different dataset characteristics. The trial of the proposed adaptive support method uses 2 basic algorithms in the association rule, namely Apriori and Fpgrowth. The test is carried out repeatedly to determine the highest and lowest minimum support values. The result showed that 6 out of 8 datasets produced minimum and maximum support values for the apriori and fpgrowth algorithms. This means that the value of the proposed adaptive support has the ability to generate a rule when viewed from the quality as adaptive support produces at a lift ratio value of > 1. The dataset characteristics obtained from the experimental results can be used as a factor to determine the minimum threshold value.

Sign in / Sign up

Export Citation Format

Share Document