**9.9 Challenges**

The major challenges that are likely to be encountered in the use of the program are the lack of staff motivation, difficulty in debugging due to failure to locate errors in coding, failure of organizations to allocate enough resources, and the practice of using machine learning language to diagnose patients for risks of heart failure.

*Enhancing Program Management with Predictive Analytics Algorithms (PAAs) DOI: http://dx.doi.org/10.5772/intechopen.98758*

**Figure 4.** *Algorithm framework for testing patients at risk of heart failure [21].*

### **9.10 Education**

Many learning institutions have used predictive analytics to predict future performances by applying past performance scores of students in their institutions. At Southern Methodist University, an associate provost has contributed to student data management practices by applying predictive analytics algorithms that combine the grades attained by students in the past years to predict their performances in the future [11].

The analysis performed involves entering the raw data into the package and following the procedure of regression analysis. The preliminary result of the regression is a regression value that is related to the current performance of the student and is a factor that enables prediction of future performance. The final outcome is a standardized coefficient that acts as a predictor of the performance of a student in future tests based on the present performance.

### **9.11 Problem statement**

The need to achieve an accurate prediction of the future performance of students at the Southern Methodist University (based on their present performances) is unquestionable. The use of a machine learning (ML) program is regarded as the most suitable approach for achieving this objective.

### **9.12 Intervention strategy**

The intervention strategy that has been recommended is the use of an ML algorithm that calculates the regression value for the students' scores, which can be used to predict their performances in the next academic periods. The recommended statistical package is GNU PSPP, which has features that enable calculation of statistical measures such as simple linear regression, multiple linear regression, cluster analysis, and reliability analysis [30].

### **9.13 Process**

The process involved was the installation of the GNU PSP application into the computer system followed by the design of the machine codes that return particular values of performance when information is input. The computer program will be composed of input points and the points of making decisions regarding the required outputs.

### **9.14 Algorithm design**

The design of the algorithm will take place immediately after the installation of the GNU PSP computer application. The design involves the use of computer decision frameworks such as the flowchart shown in **Figure 5**.

### **9.15 Pre-implementation testing**

During the pre-implementation stage, the program is tested to determine whether there are any errors. Debugging is done to correct syntax errors and factors contributing to the failure of the program are examined. The ability of the program to be used in a particular system is tested.

*Enhancing Program Management with Predictive Analytics Algorithms (PAAs) DOI: http://dx.doi.org/10.5772/intechopen.98758*

**Figure 5.** *Design of the algorithm for prediction of future student's performances [10].*
