2 Results
The total number of students who responded to the questions asked at the beginning of the semester regarding the use of Hypothes.is tool, was 64.21%. (See Table1.) The correlation between the final test results and questions asked is presented (see Figures1, 2, and3).
There were several challenges when interpreting the data by using the machine learning RapidMiner due to inconsistency in student participation during the online classes. Although students discussed and presented their views when they answered questions, not all of them wrote or participated actively during the online classes. Since the low level of online participation is not discussed here, the results are presented as a correlation between active participation (synchronous classes) and final test results (see Figure 6).
What we can observe from the machine learning data interpretation is that students who participated in weeks 3, 5, and 6 achieved a higher grade in the test. The topics in those weeks were Industry 4.0 and Quality Tools (see Figure 6 and Table 5).
Table 5
K-mean Centroid Table
| Cluster | Final exam | Week 10 | Week 12 | Week 13 | Week 3 | Week 5 | Week 6 | Week 7 | Week 9 |
|---|---|---|---|---|---|---|---|---|---|
| Cluster 0 | 52.915 | 0.630 | 0.074 | 0.185 | 1.593 | 1.111 | 1.704 | 0.741 | 0.852 |
| Cluster 1 | 55.143 | 0.138 | 0.034 | 0.034 | 1.655 | 0.724 | 0.586 | 0.138 | 0.276 |
| Cluster 2 | 37.928 | 0.148 | 0.037 | 0.074 | 0.667 | 0.630 | 0.593 | 0.296 | 0.333 |
| Cluster 3 | 52.585 | 1.167 | 0.833 | 1.500 | 1.583 | 0.750 | 1.250 | 0.750 | 1.167 |
Note: There are 4 K-mean clusters and Cluster 1 has the Final exam higher than 55.143. A strong correlation is seen in Week 3, 5, and 6 as the K-means values are the largest as 1.655, 0.724, and 0.724.
Not surprisingly, as shown in Figure 4, online class attendance decreased as the course reached the final weeks. The results could easily be correlated with the assignment results and the intensive work that students had to perform in other classes.
The reason for using the clustering technique in Machine Learning was to determine the specific properties as the following explanation is available:
Clustering is a machine learning technique that involves the grouping of data points. Given a set of data points, we can use a clustering algorithm to classify each data point into a specific group. In theory, data points that are in the same group should have similar properties and/or features, while data points in different groups should have highly dissimilar properties and/or features. Clustering is a method of unsupervised learning and is a common technique for statistical data analysis used in many fields. (Seif, 2018).
There are 4 K-clusters used in this study considered as per the following information:
K-means clustering is the most commonly used unsupervised machine learning algorithm for partitioning a given data set into a set of k groups (i.e. k clusters), where k represents the number of groups pre-specified by the analyst. It classifies objects in multiple groups (i.e., clusters), such that objects within the same cluster are as similar as possible (i.e., high intra-class similarity), whereas objects from different clusters are as dissimilar as possible (i.e., low inter-class similarity). In k-means clustering, each cluster is represented by its center (i.e., centroid) which corresponds to the mean of points assigned to the cluster (Sharma, 2019).
Reviewing data, we can see that a higher grade was achieved by students who directly participated by using the Hypothes.is tool during the classes in Weeks 3, 5, and 6 (see Table 5 and Figure 5).