Scribd
Upload
12 views4 pages

Breast Cancer Diagnosis via Data Mining

The article analyzes the use of clustering data mining techniques for the early diagnosis of breast cancer, emphasizing the IMRD structure for clarity. It discusses various algorithms such as k-mean and FF clustering, highlighting their effectiveness in diagnosing patients' health conditions. The research aims to improve health awareness and provide insights for both medical professionals and students in data mining.

Uploaded by

api-438127639
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
12 views4 pages

Breast Cancer Diagnosis via Data Mining

The article analyzes the use of clustering data mining techniques for the early diagnosis of breast cancer, emphasizing the IMRD structure for clarity. It discusses various algorithms such as k-mean and FF clustering, highlighting their effectiveness in diagnosing patients' health conditions. The research aims to improve health awareness and provide insights for both medical professionals and students in data mining.

Uploaded by

api-438127639
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

REDDIVARI SAI SARAN

DR P.J MOORE

EAP506

April 18th, 2018

ARTICLE ANALYSIS

The purpose of the article analysis is to analysis the journal articles related to my major field, to

understand the structure of the article, to identify the writing style of the author, and to know how

the research had done in my discipline by using various methods and techniques. Addition to that

we can also know how the standards of writing a research paper in my discipline differ from

writing a research paper in another discipline.

The journal article “ Diagnosis of Breast Cancer using Clustering Data Mining Approach.” is about

the diagnosis the breast cancer patient at the early stage by using some techniques [1]. The overall

structure of the article is following the IMRD structure which includes Introduction, Method,

Result, and Discussion in sequential order [2]. Most of the researcher uses IMRD structure for

writing the journal article to make audience understandable. In some articles, result and discussion

will appear in the same paper [2]. The IMRD structure provides the basic format of broad-narrow-

broad [2] which helps the authors to explain their ideas in the wider way.

From the introduction, this article explains how to solve the problem of breast cancer and early

diagnosis of the breast cancer by applying clustering data mining technique to breast cancer patient

to find the patient health condition [1]. The author then gave the literature review [1] how this

study focuses on different parameters on helping the patient to be healthy. Furthermore, the
researcher then describes the different methods for diagnosis patient’s health with different

algorithms. The researcher also demonstrates the result analysis [1] of the data by using the

algorithms in the Evaluation section. The researcher has discussed how effective algorithm works

on the patient’s health condition and the researcher also added their future work by applying

different classification and association mining algorithm [1].

The object of study is the diagnosis of breast cancer using clustering data mining [1]. The

researcher wants to diagnosis the breast cancer patient at the early stage to make the patient

healthy. On the other hand, Move 1 [2] is used to build the major information and the effect of

breast cancer in women. Additional to that the researcher explained how the breast cancer affects

women and how it can be detected.

The author’s exigence is also the Move 2 [2] which explains the challenge, problem or issue of the

previous researchers. The previous researchers do not explain how the breast cancer affect the

younger without wearing the proper corset in the ruler area [1] because it is also one of the causes

of the breast cancer. The main focus of this article is analyzing breast cancer data and the

challenging problem they use a clustering method find cancer in breast cancer dataset [1].

Move 3 [2] includes the purpose of the article is to solve the breast cancer by applying the

clustering data technique for finding the health of the patient [1] at the early stage of the breast

cancer. The article is relevant to the doctors who are specialist in treating cancer, and the women

who are suffering from the breast cancer. This article [1] also relevant to the students who are

related to the data mining, and data analytics because how the data has used to solving the breast

cancer and making the patient healthy by using some mining techniques.
The article [1] uses the IEEE (Institute of Electrical and Electronics Engineers) citation style which

is the unique style to write the technological articles. Also, this article includes tables and charts

to help the reader to understand the data visually.

SUMMARY:

Breast cancer is one of the major diseases which is found in women. Generally, breast cancer is

caused by gene changes [1]. Chest x-ray, CT scan, BONE scan and PET scan are used to detect

the stages of the breast cancer [1]. This article is used to diagnosis the breast cancer patient by

using clustering data mining technique [1]at the early stage of the patient. The author of the paper

focused on to improve health awareness, diagnosis breast cancer, the risk factor of the breast

cancer, and analyzing breast cancer dataset. To solve the problem of the breast cancer at the early

stage apply clustering data mining techniques with the algorithms like FF(Farthest First) method,

EM(Expectation Maximization) method, HCM(Hierarchical Clusterer Method) method, and k-

mean method. Each algorithm used to diagnosis’s the patient health condition. The author did the

implementation using the algorithm to calculate the healthy and sick patient. At the end of the

research k-mean and FF clustering algorithm are more helpful than the other two algorithms to

diagnosis the breast cancer of the patient [1]. The HCM algorithm having the highest error rate

and the EM algorithm cannot be able to diagnosis. To get the optimum outcome instead of using

four algorithms the author suggested using Orange, Tavera, and Rapid Miner mining tool [1].
Result Analysis:

References

[1] J. Joshi, D. Rinal and P. Jigar, "Diagnosis of Breast Cancer Using Clustering Data Mining Approach,"
International Journal of Computer Applications, vol. 101, no. 10, pp. 13-17, September 2014.

[2] J. M. Swales and F. M. Christine, Academic Writing for Graduate Student, 3rd ed., UNIVERSITY OF
MICHIGAN PRESS, 2012.

Common questions

Powered by AI

The article applies clustering data mining techniques to diagnose breast cancer at early stages by utilizing various algorithms to analyze patient health data. The algorithms discussed are FF (Farthest First), EM (Expectation Maximization), HCM (Hierarchical Cluster Method), and k-means. Among these, the k-means and FF clustering algorithms are highlighted as more effective for diagnosing breast cancer compared to the others, whereas the HCM algorithm is noted for having the highest error rate, and the EM algorithm was unable to perform satisfactory diagnosis .

The article includes tables and charts to visually represent the data and enhance the understanding of the research findings. These visual aids help readers to grasp complex information more easily and facilitate clearer comparisons and interpretations of the clustering data analysis results relevant to breast cancer diagnosis .

The article is crafted to be relevant to both medical professionals, particularly those specializing in cancer treatment, and students in data mining and analytics fields. For medical professionals, it highlights the diagnostic applications and implications of data mining techniques in breast cancer. For students, it provides a practical example of how clustering data mining techniques can be used in a healthcare context, highlighting specific algorithms and their comparative effectiveness. Additionally, the use of IEEE citation style underscores its technological orientation .

The study addresses early-stage breast cancer diagnosis by employing clustering data mining techniques to identify patient health conditions. It acknowledges previous research challenges, such as a lack of focus on how breast cancer affects younger women without proper corset usage in rural areas, which is cited as a key problem not thoroughly examined in past studies. The research fills this gap by offering a method to enhance early detection through advanced data analysis techniques .

The article acknowledges the limitations of certain algorithms, such as the significant error rate of the HCM algorithm and the poor diagnostic performance of the EM algorithm. It suggests overcoming these limitations by incorporating other mining tools like Orange, Tavera, and Rapid Miner, which may offer better functionality and more accurate diagnostic outcomes. The recommendation hints at combining different techniques to harness their individual strengths for improved accuracy in early breast cancer detection .

The article implies that clustering data mining techniques offer a computationally efficient approach to identifying early-stage breast cancer, potentially overcoming some limitations of traditional diagnostic methods that rely heavily on imaging. The data mining techniques, particularly k-means and FF algorithms, enable rapid analysis of large datasets to assess patient health, thus providing a supplementary tool that could complement traditional methods like PET scans and X-rays, which are more descriptive and diagnostic in nature but may not efficiently handle big data analytics .

The application of clustering data mining techniques in the article sits at the intersection of healthcare and data analytics. Within healthcare, these methods are used to improve diagnosis procedures for diseases like breast cancer by analyzing complex patient datasets. From a data analytics perspective, the use of clustering algorithms and mining techniques showcases the capacity to manage large volumes of data to extract meaningful patterns and insights. Thus, the research embodies a multidisciplinary approach to solving medical challenges by leveraging computational tools and methodologies .

The article utilizes the IMRD structure, which stands for Introduction, Method, Results, and Discussion, to organize and convey scientific information logically and sequentially. This structure helps by providing a broad-narrow-broad format, which allows the author to introduce the topic broadly, narrow down to specific methodologies and results before broadening out again in the discussion section to relate findings to the wider field. This approach aids in making the research more understandable and accessible to readers .

The study identifies genetic changes as a primary factor in causing breast cancer. It evaluates these changes and the stages of breast cancer using diagnostic tools such as chest X-ray, CT scan, BONE scan, and PET scan to determine the progression and stage of the disease. These traditional diagnostic tools complement the data mining techniques discussed in the article for comprehensive diagnosis .

The article suggests future research directions towards implementing more efficient data mining tools such as Orange, Tavera, and Rapid Miner for optimized outcomes, instead of relying solely on the four discussed algorithms. This indicates a shift towards hybrid or more sophisticated analytics tools to improve diagnostic accuracy and suggests expanding algorithmic exploration beyond clustering methods to include classification and association mining techniques, thereby enhancing early diagnostic capabilities .

You might also like