Scribd
Upload
12 views12 pages

Presentation 9

Presentation CNN Deep learning

Uploaded by

Bello Mada
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
12 views12 pages

Presentation 9

Presentation CNN Deep learning

Uploaded by

Bello Mada
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 12

A background study on

Anomaly-based Intrusion Detection Approach for IoT


Networks Using Machine Learning
Presented
By

AHMAD, Bello
20210310014
1
Implementation of the Anchor paper
• Weka is a collection of machine learning algorithms and data
preprocessing tools for data mining tasks.

• It contains tools for data preparation, classification, regression, clustering,


association rules mining, and visualization.

2
Dividing the dataset in subsets
• From the Anchor, three subsets were drive from the IoTID20 dataset
• Subset 1  Normal and DoS
• Subset 2  Normal and MITM
• Subset 3  Normal and Scan

• To achieve this, filters are use

• The weka.filters package is concerned with classes that transform datasets by


removing or adding attributes, resampling the dataset and so on.

SubsetByExpression which filters instances according to a user-specified expression.


(CLASS is ‘Normal') or (CLASS is ‘DoS')

3
Preprocess the subsets
• Each subset will be preprocess separately in order remove unwanted
instances and attribute for dimensionality reduction.

RemoveUseless Filter will be use


• This filter removes attributes that do not vary at all or that vary too much.

StringToNominal filter
• Converts a range of string attributes (unspecified number of values) to
nominal (set number of values).

NominalToBinary
• Converts all nominal attributes into binary numeric attributes. An attribute 4
Preprocess the subsets (Cont…)
Normalize
• Normalizes all numeric values in the given dataset (apart from the class
attribute, if set).

• OR

Standardize
• Standardizes all numeric attributes in the given dataset to have zero mean and
unit variance (apart from the class attribute, if set).

5
Attribute selection (Feature selection )
• Click the “Select attributes” tab to access the feature selection methods.
• Attribute Evaluator
• Search Method.

• The attribute evaluator is the technique by which each attribute in your


dataset is evaluated in the context of the output variable.

• The search method is the technique by which to try or navigate different


combinations of attributes in the dataset in order to arrive on a short list
of chosen features.

6
Attribute selection (Feature selection ) (Cont…)

7
Attribute selection (Feature selection ) (Cont…)

8
Attribute selection (Feature selection ) (Cont…)
• Subset 1
Subset Anchor (SF) This work(SF)
1 15 13
2 11 12

3 11 13

9
Classify
• Weka provides a range of classifiers that are suitable for different purposes
and scenarios.

10
Result obtained

Subset Anchor(Accuracy) This work(Accuracy)


1 99.95% 100%
2 99.98% Process terminated
3 99.96% Process terminated

11
Thank you

12

You might also like