Detecting spammers on social networks

In the citing paper

“Such feature sets are then used to train and test machine learning classifiers in order to distinguish between humans and spammers.”

In the citing paper

“Both of them propose a list of features to be extracted from manually classified datasets of accounts.”

In the citing paper

“With both the [2] and [8] feature sets, BN, kNN and LR classifiers achieve, overall, worse performances.”

In the citing paper

“3. the content of spambots' tweets exhibits the so-called message similarity ;”

In the citing paper

“The study presented in [8] focuses on spambot detection.”

In the citing paper

“The ones built over the feature set of [2] achieve slightly better results.”

In the citing paper

“1. spambots do not have thousands of friends;”

In the citing paper

“In addition, all the classifiers based on the feature set by [2] have a higher AUC , when compared with the ones built with the feature set by [8].”

In the citing paper

“Anyway, the latter also obtain high detection levels: accuracy, precision, and recall are around 0.98 for RF, D and J48, with a MCC of around 0.96.”

In the citing paper

“The lower precision and recall with respect to the ones obtained using the set of Yang et al. [2] show that the features of Stringhini et al. [8] exhibit the tendency to consider as fake followers some human accounts.”

In the citing paper

“total tweets 5. spambots have a high friends ( followersˆ2 ) ratio value (i.e., lower ratio values mean legitimate users).”

In the citing paper

“The authors exploit several characteristics that can be gathered crawling an account's details, both from its profile and timeline. For each investigated account, such characteristics are exploited in a Random Forest algorithm [23, 43], that outputs if the account is a spambot or not.”

In the citing paper

“Even if the proposed features have been originally designed for spam detection, here, for the first time, we consider them to spot another category of Twitter accounts, i.e., the fake followers.”

In the citing paper

“4. spambots have a high tweets containing URLs ratio ( URL ratio );”

In the citing paper

“2. spambots have sent less than 20 tweets;”

In the citing paper

“We start observing that the Class C classifiers, built over all the features considered in our study, perform better than all the other ones, including the classifiers using the feature sets of Yang et al. [2] and Stringhini et al. [8], as reported in Table 12.”

In the citing paper

“meaning that the latter is able to miss fewer fakes than the former one after pruning.”

In the citing paper

“The decision tree obtained with the feature set of Stringhini et al. [8] has 22 leaves, 43 nodes, and a height of 7, whereas the best feature is the friends/(followersˆ2) ratio that places at the root.”

In the citing paper

“Looking at the tree structure, we observe some interesting differences between the fake followers in our BAS dataset and the spam accounts characterized in [8] and [2].”

In the citing paper

“Furthermore, we clearly observe that the pruned version of Stringhini et al. [8] reduces its recall of 0.017, while the one of Yang et al. [2] only drops of 0.004”

In the citing paper

“Overall, the results of this experiment show that, even reducing the features, it is possible to have a detection rate higher than 0.95 (as in the last lines of Table 13, for [8] and [2], respectively).”

In the citing paper

“Similarly, we have seen that the feature set proposed by Yang et al. [2] seems to be slightly more effective than the one proposed by Stringhini et al. [8], when used in feature-based classifiers aiming at fake followers detection.”

In the citing paper

“Unlike [30, 44], we leverage timeline-based features for the detection of fake accounts.”