Real time data analysis and visualization for the breast cancer disease

Abstract

Today, the amount of data that are digitally collected in the healthcare sector is tremendous and expanding rapidly, these data are inherently geospatial and temporal ranging from individual families to whole states and from minutes to decades. Therefore, they need sophisticated data management and analysis to be transformed into valuable knowledge. Healthcare professionals are faced with several challenges regarding extracting knowledge from this massive amount of data in order to support the decision-making process. To gain advantage of health care big data, big data analytics need to be exploited to utilize and understand patterns associations within these data thus make the right decision. In this research, an interactive data analysis and visualization tool is proposed to visually compare the performance of three machine learning algorithms on Wisconsin Diagnostic Breast Cancer (WDBC) dataset. The proposed model consists of two phases: input phase and analysis/visualization phase. It aims to allow the user to interactively compare the performance of three different ML algorithms (KNN, SVM and NB) in terms of accuracy, sensitivity and error rate in a user-friendly way. Here, SVM classifier has proven its efficiency and it is concluded as the best classifier with the highest accuracy as compared to the other two classifiers.

Breast Cancer Detection by Using Data Mining, a Review Study

“…An interactive data analysis and visualization tool is proposed [19]. Visualizations were intended to compare the performance of three machine learning algorithms applied on the Wisconsin Diagnostic Breast Cancer (WDBC) dataset.…”

“…Naïve Bayes data mining technique was used alone or with other techniques as in [12], [14], [15], and [16]. Researchers used Support Vector Machine (SVM) with other techniques in [17], [18], and [19]. Regarding the results, the confusion matrix was used alone or with other evaluation techniques as in [13], [15], [17], and [19].…”

“…Researchers used Support Vector Machine (SVM) with other techniques in [17], [18], and [19]. Regarding the results, the confusion matrix was used alone or with other evaluation techniques as in [13], [15], [17], and [19]. Other evaluation techniques were used, like benchmark data set in [12], visual evaluation in [16], and comparative analysis in [20].…”

“…Other evaluation techniques were used, like benchmark data set in [12], visual evaluation in [16], and comparative analysis in [20]. Concerning data sets used, Wisconsin data set was used in [17], and [19]. The Surveillance Epidemiology and End Results (SEER) data set was used in [14] and [18].…”

Real-Time Machine Learning-Based Framework for the Analysis of Banking Financial Data

“…Today, with enormous amounts of information produced each day, by having many different sources of data collection, it has gotten fundamental for business and financial sector to proficiently oversee and process their insight. Therefore, it will be able to offer best types of assistance, achieve operational greatness and upgrade the basic leadership process, through making the right decisions in real time [5].…”