Detecting network attacks in information security using machine learning techniques
| dc.contributor.advisor | Kurnaz, Sefer | |
| dc.contributor.author | Ahmed, Ali Saadoon Ahmed | |
| dc.date.accessioned | 2023-07-28T08:35:55Z | |
| dc.date.available | 2023-07-28T08:35:55Z | |
| dc.date.issued | 2022 | en_US |
| dc.date.submitted | 2022 | |
| dc.department | Enstitüler, Lisansüstü Eğitim Enstitüsü, Bilişim Teknolojileri Ana Bilim Dalı | en_US |
| dc.description.abstract | DDoS attacks are usually ranked higher than other networks and IDS threats (IDS). Customers that seek services that are dependent on the servers in question will experience service delays, resulting in monetary and reputational losses for the company. in this thesis, we applied eight defiant machine learning approaches to distinguish between two distinct attacks. DoS AttacksHulk and DoS Attacks-SlowHTTPTest were the names of these attacks. The algorithms DT, RF, GB, AdaBoost, NB, XGB, and Ridge are some examples of MLP is one more. The CICIDS2017 data set was utilized as the basis for analysis. This dataset was located in the Information Security section of Kaggle. This dataset includes millions of cases, each representing one of 80 attributes across 15 distinct attacks. By Using Python Environment version 3.10 we applied these methods to the dataset's DoS attacks to categorize them as DoS or benign attacks in two distinct Experiments: A first experiment is Binary Classification Experiment in short, we will apply and train the eight ML techniques used in this study this Experiment between normal and malicious (Benign and DoS Attack). Second Experience is Multiclass Classification Experiments, In this Experiment two malicious attack types (DoS Attacks-Hulk) and (DoS Attacks-Slow HTTP Test). The purpose of these Experiments is to propose a detection model based on ML algorithms that can enhance the Accuracy reported in previous works. in this thesis, we applied two different experiments, and we obtained terrific results: in the first experiment, we had a success rate of over 99% and in the second we had a 100% success rate, the results of this thesis are more efficient compared to the results of previous work used in Chapter2 of this thesis. | en_US |
| dc.identifier.citation | Ahmed, Ali Saadoon Ahmed. (2022). Detecting network attacks in information security using machine learning techniques . (Yayınlanmamış yüksek lisans tezi). Altınbaş Üniversitesi, Lisansüstü Eğitim Enstitüsü, İstanbul. | en_US |
| dc.identifier.uri | https://hdl.handle.net/20.500.12939/3610 | |
| dc.identifier.yoktezid | 796231 | |
| dc.institutionauthor | Ahmed, Ali Saadoon Ahmed | |
| dc.language.iso | en | |
| dc.publisher | Altınbaş Üniversitesi / Lisansüstü Eğitim Enstitüsü | en_US |
| dc.relation.publicationcategory | Tez | en_US |
| dc.rights | info:eu-repo/semantics/closedAccess | en_US |
| dc.subject | Network Attacks | en_US |
| dc.subject | Information Security | en_US |
| dc.subject | Machine Learning | en_US |
| dc.subject | CICIDS2017 | en_US |
| dc.subject | DoS attacks | en_US |
| dc.title | Detecting network attacks in information security using machine learning techniques | |
| dc.type | Master Thesis |
Dosyalar
Lisans paketi
1 - 1 / 1
[ X ]
- İsim:
- license.txt
- Boyut:
- 1.44 KB
- Biçim:
- Item-specific license agreed upon to submission
- Açıklama:
