See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/332173637

Prediction of Breast Cancer Disease Using Decision Tree Algorithm

Article · April 2019

CITATION READS

1 620

3 authors:

Nathan Nwokocha Ledisi G. Kabari

Rivers State University of Education Ken Saro-Wiwa Polytechnic, Bori
3 PUBLICATIONS   1 CITATION    60 PUBLICATIONS   100 CITATIONS   

SEE PROFILE SEE PROFILE

Francis Agaba
University of Calabar
5 PUBLICATIONS   1 CITATION   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Telecommunications Subscription Fraud Detection View project

Present Day Internet Design, Architecture, Performance and an Improved Design View project

All content following this page was uploaded by Ledisi G. Kabari on 03 April 2019.

The user has requested enhancement of the downloaded file.

 International Journal of Innovative Information Systems
& Technology Research 7(1):34-38, Jan.-Mar., 2019

© SEAHI PUBLICATIONS, 2019 www.seahipaj.org ISSN: 2467-8562

Prediction of Breast Cancer Disease Using Decision Tree

Algorithm

1
Nwokocha, Nathan, 2Ledisi Kabari & 3Agaba, Francis
1
Department of Computer Science Education
Federal College of Education (Technical) Omoku, Rivers State, Nigeria
2
Ken Saro-Wiwa Polytechnics, Bori, Rivers State, Nigeria

ABSTRACT
The Healthcare industry is generally “information rich”, but unfortunately not all the data mined can
be used for discovering hidden patterns and effective decision making. Advanced data mining
techniques are used to discover knowledge in database and for medical research, particularly in Heart
disease prediction. This paper has analysed prediction systems for Breast Cancer disease using
Decision tree algorithm and WEKA 3.8 as a machine learning tool. The system uses medical terms
such as Age, menopause, tumour size, inv-nodes, irradiant, node- caps, deg-malig, breast, breast- quad
and class. These are the 10 attributes including class applied to predict the likelihood of patient having
a breast cancer disease. The class is either no recurrence event or recurrence event.
Keywords: Breast cancer, Classification, Prediction, Decision trees.

INTRODUCTION
According to WHO (2002) cancer has been responsible for the death of millions of people
Worldwide, with an estimated increase of 50% for developing countries and 70% of the total death are
due to cancer. According to Parkin, Ferlay, Hamdi-Cherif, Sitas, Thomas and Wabinga (2003)
developing nations only possess 5% of global funds for cancer control. It is opined that very few
human and material resources are available in these countries (Grey & Sener, 2006).
Breast cancer is a malignant tumour (a collection of cancer cells) arising from the cells of the breast.
Although breast cancer predominantly occurs in women, it can also affect men. This article deals with
breast cancer in women. Breast cancer and its complications can affect nearly every part of the body.
Breast cancer does not always produce symptoms; women may have cancers that are so small which
do not produce masses and can be felt or by other recognizable changes in the breast. When
symptoms do occur, a lump or mass in the breast is the most common symptom.
Other possible symptoms include
 nipple discharge or redness,
 changes in the skin such as puckering or dimpling,
 Swelling of part of the breast.
In machine learning and statistics, classification is a supervised learning approach in which the
computer program learns from the data input given to it and then uses this learning to classify new
observation. This data set may simply be bi-class (like identifying whether the person is male or female
or that the mail is spam or non-spam) or it may be multi-class too. Some examples of classification
problems are: speech recognition, handwriting recognition, bio metric identification, document
classification etc.
This study aims at using data mining techniques to classify breast cancer risks using datasets of
patients’ information from UCI Repository which contains the risk factors and the cancer classes. The
decision trees classification of breast cancer was performed using the WEKA software.

34
 Nwokocha et al. ….. Int. J. Innovative Info. Systems & Tech. Res. 7 (1):34-38, 2019

Literature Review
Rajesh and Anand (2012) employed SEER dataset for the diagnosis of breast cancer using the C4.5
classification algorithm. The algorithm was used to classify patients into either pre-cancer stage or
potential breast cancer cases. Random tests were performed on the dataset which contained
information for 1183 patients including the age of diagnosis, regional lymph nodes measures, and
sequence number of tumours, dimension of primary tumours and contiguous growth of the primary
tumours. The analysis involved the use of three random 500 records form the pre-processed data of
1183 and was used as training data and the lowest error rate achieved was 0.599. During the testing
phase, the C4.5 classification rules were applied to a test sample and the algorithm showed had an
accuracy of 92.2%, sensitivity of 46.66% and a specificity of 97.4%. Future enhancement of the work
will require the improvisation of the C4.5 algorithm to improve classification rate to achieve greater
accuracy.
Shajahan et al (2013) worked on the application of data mining techniques to model breast cancer data
using decision trees to predict the presence of cancer. Data collected contained 699 instances (patient
records) with 10 attributes and the output class as either benign or malignant. Input used contained
sample code number, clump thickness, cell size and shape uniformity, cell growth and other results
physical examination. The results of the supervised learning algorithm applied showed that the
random tree algorithm had the highest accuracy of 100% and error rate of 0 ,while CART had the
lowest accuracy with a value of 92.99% but naïve bayes’ had the an accuracy of 97.42% with an error
rate of 0.0258.
Mangasarian, Street and Wolberg (1995) performed classification on both diagnostic and prognostic
breast cancer data. The classification procedure adopted by them for diagnostic data is called Multi
Surface Method-Tree (MSM-T) .This used a linear programming model to iteratively place a series of
separating planes in the feature space of the examples. The procedure is recursively repeated.
Moreover they have approached the prognostic data using Recurrence Surface Approximation (RSA)
which used linear programming to determine a linear combination of the input features which
accurately predicts the Time-To-Recur (TTR) for a recurrent breast cancer case. The training
separation and the prediction accuracy with the MSM-T approach was 97.3% and 97 % respectively,
whereas the RSA approach was able to give accurate prediction only for each individual patient. Their
drawback was the inherent linearity of the predictive models.
Data Mining Process
Data mining is defined as a process of discovering hidden valuable knowledge by analysing large
amounts of data which is stored in databases or data warehouse using various data mining
techniques such as machine learning, artificial intelligence(AI) and statistical. The two basic data
mining tasks are: descriptive data mining tasks which help to understand the characteristic properties
of dataset and predictive data mining tasks which are used to perform predictions based on available
dataset. Predictive data mining is the chosen data mining task for this study.

METHODOLOGY
The breast cancer disease dataset has been taken from the UCI Repository,
https://archive.ics.uci.edu/ml/machine-learning. This dataset has 286 instances and 10 attributes. The
attributes are:
 Age,
 Menopause,
 Tumour size,
 Inv-nodes,
 Irradiant,
 Node- caps,
 Deg- malig,
 Breast,
 Breast- quad
 Class

35
 Nwokocha et al. ….. Int. J. Innovative Info. Systems & Tech. Res. 7 (1):34-38, 2019

Data mining Algorithm used

Decision Trees
Decision tree builds classification or regression models in the form of a tree structure. It breaks down a
data set into smaller and smaller subsets, while at the same time an associated decision tree is
incrementally developed. The final result is a tree with decision nodes and leaf nodes. A decision node
has two or more branches and a leaf node represents a classification or decision. The topmost decision
node in a tree which corresponds to the best predictor is called root node. Decision trees can handle
both categorical and numerical data.
Performance Evaluation
The performance evaluation criteria allow the measurement of the accuracy of the model developed
using the training dataset. The results of the classification are recorded on a confusion matrix. A
confusion matrix is a square which shows the actual classification along the vertical and the predicted
along the vertical. All correct classifications lie along the diagonal from the north-west corner to the
south-east corner, also called True Positives (TP) and True Negatives (TN) .The other cells are called
the False Positives (FP) and False Negatives (FN).
Experimental Result
The experimental results of this study used the classifier and the WEKA software data mining tool
was employed. In the experiment, two classes were used and therefore a 2x2 confusion matrix was
applied. Class a = No recurrence event and Class b = Recurrence event. The result obtained from use
of the WEKA software is displayed in fig 5. The result summary of the classifier output is shown in
Figure 6.Table 1 indicates the accuracy of the result in percentage using the Decision Tree, while fig 7
shows the result in pie-chart.

Figure 1: Graphical representation of attribute

36
 Nwokocha et al. ….. Int. J. Innovative Info. Systems & Tech. Res. 7 (1):34-38, 2019

Figure 2: Result Summary of classifier output

Table 1: Accuracy Table
Algorithm Correctly Incorrectly
Classified Classified
Instance Instance (%)
(%)
Decision 66% 34%
Tree

37
 Nwokocha et al. ….. Int. J. Innovative Info. Systems & Tech. Res. 7 (1):34-38, 2019

Figure 3: Accuracy, Correct and incorrect classification of Breast Cancer

CONCLUSION
In this study a data mining classification technique was used for the prediction of breast cancer
disease. Experimental result shows that the correctly classified instances for breast cancer are greater
than the incorrectly classified instances considering 286 instances and 10 attribute.

REFERENCES
Grey, N & Sener, S. (2006). Reducing the global cancer burden Retrieved 5th November, 2018 from
http://www.hospitalmanagement.net/features/feature648/,
Mangasarian, D. S. Street, W.N. & Wolberg, W.H. (1995). Breast cancer diagnosis and prognosis via
linear programming, Operations Research, 43(4), 570-577.
Parkin, D.M., Ferlay, J., Hamdi-Cherif, M., Sitas, F., Thomas, J.O., Wabinga, H., Whelan, S.L.
(2003). Cancer in Africa Epidemiology and Prevention, IARC (WHO) Scientific Publications
no. 153, IARC Press, Lyon, France.
Rajesh, K. & Anand, S. (2012). Analysis of SEER dataset for breast cancer diagnosis using C4.5
classification algorithm. International Journal of Advanced Research in Computer and
Communication Engineering, 1(2), 72 – 77.
Shajahaan, S.S., Shanthi, S. & ManoChitra, V. (2013). Application of data mining techniques to
model breast cancer data. International Journal of Emerging Technology and Advanced,
3(11), 3622-369
World Health Organization (2002). National Cancer Control Programmes; policies and managerial
guidelines, 2nd edition.

38

View publication stats