The Growing Threat of Phishing and Smishing

Oct 13, 2025 by recoveryshake.com 64 views

The Growing Threat of Phishing and Smishing

Phishing is a type of crime on social media that aims to steal personal data from individuals and industries. One form of phishing that is now widespread is through text messages, known as smishing. Smishing messages often contain email addresses, telephone numbers, or website links that are designed to attract the attention of the recipient without them knowing the threatening dangers. This kind of crime can harm victims financially and endanger the security of their data. Even so, the identification of Smishing is still a big challenge, because at this time the process is still carried out manually, which requires a long time. Therefore, a approach is needed that can detect phishing on text messages faster and more accurately.

The increasing number of smishing attacks has made it essential to develop a system that can identify and filter out these malicious messages. With the rise of mobile devices and the widespread use of text messaging, smishing has become a significant threat to individuals and organizations. The consequences of falling victim to smishing can be severe, including financial loss, identity theft, and damage to reputation. Therefore, it is crucial to develop a system that can detect smishing messages accurately and efficiently.

The Support Vector Machine Algorithm and Ensembled Bagging

This study aims to identify phishing in text messages using the Support Vector Machine (SVM) algorithm combined with the Ensembled Bagging technique. The SVM algorithm is a popular machine learning algorithm that is widely used for classification tasks. It works by separating data into two classes based on existing features. The Ensembled Bagging technique is a method of improving the accuracy of the SVM algorithm by combining the predictions of multiple models.

The methodology used in this study includes data collection from 1600 messages, both in the phishing and non-phishing categories, obtained from previous studies and research boxes. After that, the data is processed through several stages, namely cleaning (cleaning), tokenisasi, the elimination of general words (stopwords elimination), and indexing basic words (stemming). The data is then divided into two sets, namely the training set and the test set.

Data Preprocessing

The preprocessing process carried out, such as cleaning and removing stopwords, is very important to improve the quality of the data used. Cleaning text can eliminate unnecessary elements and focus on more relevant information. The use of tokenisation helps in breaking sentences into smaller words, thus facilitating further analysis. Meanwhile, the removal of stopwords helps to reduce the size of the processed data, without losing the important meaning of the text.

The cleaning process involves removing special characters, punctuation, and numbers from the text data. This is done to eliminate unnecessary elements and focus on the relevant information. Tokenisation is then applied to break down the text into smaller words or tokens. This helps to facilitate further analysis and improves the accuracy of the SVM algorithm.

The Ensembled Bagging Technique

The Ensembled Bagging technique involves random sampling from training data and training several SVM models simultaneously. The end result is a combination of these models, which can reduce variance and increase prediction stability. This technique is particularly useful when dealing with high-dimensional data, where the number of features is large compared to the number of samples.

The Ensembled Bagging technique works by creating multiple copies of the training data and training an SVM model on each copy. The predictions of each model are then combined to produce a final prediction. This technique can improve the accuracy of the SVM algorithm by reducing overfitting and increasing the robustness of the model.

Results and Discussion

The results of this study show that the proposed approach can identify phishing with accuracy reaching 95.2%. This proves that the combination of SVM and Ensembled Bagging is very effective in processing text messages data for phishing identification purposes.

The accuracy of the proposed approach is significantly higher than the accuracy of the SVM algorithm alone. This is due to the ability of the Ensembled Bagging technique to reduce variance and increase prediction stability. The results of this study demonstrate the effectiveness of the proposed approach in identifying phishing messages.

Conclusion and Future Work

Future work can involve improving the accuracy of the proposed approach by using more advanced machine learning algorithms or by incorporating additional features into the model. Additionally, the proposed approach can be applied to other types of text data, such as emails or social media posts, to identify phishing messages.

Implications and Recommendations

The results of this study have significant implications for the development of systems that can detect and filter out phishing messages. The proposed approach can be used to improve the accuracy of existing systems and to develop new systems that can detect phishing messages more effectively.

The study also highlights the importance of awareness and education in preventing phishing attacks. Individuals and organizations must be aware of the potential threats from phishing and take steps to protect themselves from these attacks.

Limitations and Future Research Directions

The study has several limitations, including the use of a small dataset and the lack of diversity in the data. Future research can involve collecting a larger and more diverse dataset to improve the accuracy of the proposed approach.

Additionally, future research can involve exploring other machine learning algorithms and techniques that can be used to improve the accuracy of the proposed approach. The study also highlights the need for further research in the area of phishing detection and prevention.

Conclusion

In conclusion, this study has demonstrated the effectiveness of the Support Vector Machine algorithm combined with the Ensembled Bagging technique in identifying phishing messages. The proposed approach has achieved an accuracy of 95.2%, which is significantly higher than the accuracy of the SVM algorithm alone. The study has significant implications for the development of systems that can detect and filter out phishing messages and highlights the importance of awareness and education in preventing phishing attacks.

Q: What is phishing and why is it a threat?

A: Phishing is a type of cybercrime that involves tricking individuals into revealing sensitive information such as passwords, credit card numbers, or personal data. It is a threat because it can lead to financial loss, identity theft, and damage to reputation.

Q: What is smishing and how is it different from phishing?

A: Smishing is a type of phishing that occurs through text messages (SMS) or multimedia messages (MMS). It is different from phishing because it uses mobile devices to send malicious messages to victims.

Q: What is the Support Vector Machine (SVM) algorithm and how does it work?

A: The SVM algorithm is a type of machine learning algorithm that is used for classification tasks. It works by separating data into two classes based on existing features. The algorithm uses a kernel function to map the data into a higher-dimensional space, where it can be separated more easily.

Q: What is Ensembled Bagging and how does it improve the accuracy of the SVM algorithm?

A: Ensembled Bagging is a technique that involves creating multiple copies of the training data and training an SVM model on each copy. The predictions of each model are then combined to produce a final prediction. This technique can improve the accuracy of the SVM algorithm by reducing overfitting and increasing the robustness of the model.

Q: What are the advantages of using the SVM algorithm with Ensembled Bagging for phishing detection?

A: The advantages of using the SVM algorithm with Ensembled Bagging for phishing detection include:

High accuracy: The algorithm can achieve high accuracy rates, even with small datasets.
Robustness: The algorithm is robust to noise and outliers in the data.
Flexibility: The algorithm can be used with a variety of features and datasets.

Q: What are the limitations of using the SVM algorithm with Ensembled Bagging for phishing detection?

A: The limitations of using the SVM algorithm with Ensembled Bagging for phishing detection include:

Computational complexity: The algorithm can be computationally expensive, especially with large datasets.
Overfitting: The algorithm can suffer from overfitting if the training data is not representative of the test data.

Q: How can the SVM algorithm with Ensembled Bagging be used for real-world phishing detection?

A: The SVM algorithm with Ensembled Bagging can be used for real-world phishing detection by:

Collecting and preprocessing data from various sources, such as email, social media, and text messages.
Training the algorithm on the preprocessed data to detect phishing messages.
Deploying the algorithm in a real-world setting, such as a spam filter or a phishing detection system.

Q: What are the future research directions for phishing detection using machine learning algorithms?

A: Future research directions for phishing detection using machine learning algorithms include:

Developing more accurate and robust algorithms for phishing detection.
Exploring the use of deep learning algorithms for phishing detection.
Investigating the use of transfer learning and domain adaptation for phishing detection.

Q: How can individuals and organizations protect themselves from phishing attacks?

A: Individuals and organizations can protect themselves from phishing attacks by:

Being aware of the potential threats from phishing.
Using strong passwords and two-factor authentication.
Being cautious when clicking on links or downloading attachments from unknown sources.
Using anti-phishing software and filters.