Fig 1.
Workflow of the ISBP model.
Fig 2.
Visualized decision tree learning sample.
The decision tree classifier will learn the regularity in users’ sharing behavior, and generate a flowchart-like structure starting from the root node, with paths connecting several leaf nodes and each representing a class label. As shown above, there are 57 users also agreed to share the item 20, in the total of 63 users who had agreed to share the item 29, so this rule is titled with 57/63 predicting accuracy.
Fig 3.
Suppose there is a dataset consisting a group of users’ sharing actions, WEKA will classify each user by a majority vote of its neighbors, assigning to the class most common among its nearest neighbors. The dataset will be randomly divided into 10 folds, and each fold will be used as the testing data while the rest will be applied as the training set. This sample regards K = 8, and the classified results indicate the precision is very high with over 87% percent of users are classified correctly.
Fig 4.
Naive Bayes Classifier Sample.
The Naïve Bayes classifier will assign class labels to each user, represented as vectors of feature value, such as all the users who agreed to share item 1–7, 13–17, 24–29, while refused to share item 8–12, 18–23 will be classified into class 1: “agree to share the item 12”. This figure shows an example of Naïve Bayes with 8 identified classes.
Fig 5.
The sharing counts for each requested item with different recipients.
Very few items were shared with strangers (triangles); more items were shared with friend (circles) and family (squares), and some items were preferred to share with nobody (stars).
Table 1.
Descriptive statistics for the 30 items answered by the users.
Fig 6.
Mean and standard deviation of compared pairs with different features.
a and b represent the compared values between CI and DI, c and d represent the compared values between male participants and female participants, e and f represent the compared values between older participants and younger participants, and g and h represent the compared values between computer major participants and non-computer major participants.
Fig 7.
Predicting accuracy for participants’ sharing decisions.
The models were built with three ISBP based ML techniques: decision tree classifier (squares), k-nearest neighbor (circles), and Naive Bayes classifier (stars).
Fig 8.
Error rates are compared in each condition.
Males (triangles) vs. females (squares), younger participants (triangles) vs. older participants (squares), computer majors (triangles) vs. non-computer majors (squares).
Fig 9.
The disclosure of users in Condition 1 and Condition 2.
Faculty members (older users) tend to behave more stably than students (younger users) in sharing the information, whereas computer majors are less varied than non-computer majors in information sharing. As a result, computer major faculty members’ behavior is most stable, and non-computer major students’ sharing behavior is the least stable.
Table 2.
Prediction accuracy for users in condition 1 and condition 2.