Scribd
Upload
148 views12 pages

Association Rules Data Mining

This document discusses data mining association rules and the Apriori algorithm. It provides examples of how association rules can be used, such as for market basket analysis or medical diagnosis. The Apriori algorithm is described as a popular method to find frequent patterns in datasets by looking at support, confidence and lift. Steps of the Apriori algorithm are outlined. The document also provides a case study about how a local official in Indonesia is using association rules and Apriori to determine which community groups should collaborate to address a COVID-19 outbreak.

Uploaded by

Ahmad Yd
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
148 views12 pages

Association Rules Data Mining

This document discusses data mining association rules and the Apriori algorithm. It provides examples of how association rules can be used, such as for market basket analysis or medical diagnosis. The Apriori algorithm is described as a popular method to find frequent patterns in datasets by looking at support, confidence and lift. Steps of the Apriori algorithm are outlined. The document also provides a case study about how a local official in Indonesia is using association rules and Apriori to determine which community groups should collaborate to address a COVID-19 outbreak.

Uploaded by

Ahmad Yd
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 12

Data Mining

(Association Rules)
Muhamad Fathurahman, M.Kom
Fakultas Teknologi Informasi
Universitas YARSI
Outline


Association Rules Application

Apriori Algorithm

Implementing Association Rules in Python
with a case study

Association Rules
Application

Association rules is one of the data-mining
task that find repeated occurence of attribute
item in a dataset
 Shopping Basket Analysis (E.g purchases,
item searches
 Medical Diagnosis (Occurrence of illness
with symptoms
 Census Data (E.g increase of public
services (Education, health, transport)

Association Rules
Application

Association Rules
Application
Association Rules Algoritm


Appriori – Most common popular algorithm to
find repeated pattern in a dataset
 Support
 Confidence
 Lift
Association Rules Algoritm

Appriori – Most common popular algorithm to find
repeated pattern in a dataset
 Support
 Ratio of an item occur in a transaction.

To calculate the support of item B

support(B) =
transaction_contained(B)/total_transaction

support(Burger) = 10/100 = 10%
Association Rules
Algorithm

Appriori – Most common popular algorithm to find
repeated pattern in a dataset
 Confidence

Calculate the likelihood if an item B is bought then
an item A is also bought
 confidence(B->A) = (trainsaction_contained(B,A)
/ transaction_contained(B)
 confidence(Burger -> Cola) = 8/10 = 80%
Association Rules Algoritm

Appriori – Most common popular algorithm to find
repeated pattern in a dataset
 Lift

Increased ratio of item B when item A is bought.

Lift(B->A) = confidence(B->A) / support(A)

Lift(Burger->Cola) = 80/20 = 4.0
Association Rules Algoritm

Apriori Algorithm – Steps
 Determined the value of mininum-support and
minimum-confidence.
 Select items or subset of items that have support
values greater than or equal to minimum-support
 Select all subset of items that have confidence value
greater than or equal to minimum-confidence
 Sort the rules that according to the lift values
Case Study: Attack Covid-19

Mang Ujang adalah ketua satgas covid-19 di kabupaten Kofu.
Pemerintah pusat telah memutuskan Kabupaten Kofu menerapkan
PSBB karena jumlah kasus aktif terus meningkat (78 kasus). Di
kabupaten Kofu, rata-rata 60 penduduknya adalah buruh harian.
Mereka termasuk penduduk yang terdampak karena kebijakan
PSBB.

Namun sebagian besar penduduk Kabupaten Kofu, adalah anggota
aktif dari banyak komunitas-komunitas yang terdapat di kabupaten
Kofu. Mang Ujang ingin menggerakan komunitas-komunitas ini
untuk saling bekerja sama dalam hal pemberantasan covid-19.

Mang Ujang ingin melihat ‘hubungan keterkaitan secara alami’
(association) dari komunitas-komunitas yang ada sebelum
memutuskan kerjasama untuk mengatasi wabah ini
Case Study: Attack Covid-19

Mang Ujang kemudian membuat sebuah survey untuk
mendapatkan data keanggotaan dari masyarakat yang aktif pada
beberapa komunitas di kabupaten Kofu.

Untuk mendapatkan ‘Knowledge’ dari survey tersebut, Mang Ujang
menggunakan algoritma Apriori untuk menentukan komunitas-
komunitas mana yang dapat bekerjasama.

You might also like