Week 2 Basheer

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

DATE: 09-02-2024

WEEK-2

AIM: Pre-process a given dataset based on the following: A. Attribute Selection


B. Handling Missing Values
A) Attribute Selection :Refers to the process of selecting a subset of relevant features from
your dataset. This can help improve the performance of machine learning models by
reducing dimensionality and focusing on the most informative attributes. WEKA provides
several built-in algorithms and tools for attribute selection.
STEPS for Attribute Selection
1. Launch the WEKA application.
2. Load Data: Go to the "Explorer" tab and load your dataset using the "Open file..."
button. Select Attribute: Navigate to the "Select attributes" tab.

3. Choose Attribute Evaluator: From the "Attribute Evaluator" dropdown menu, select
an algorithm for evaluating attribute.
4. Select Search Method: Choose a search method from the dropdown menu.

B) Handling Missing Values


Identify Missing Values: Before proceeding with any analysis, identify the missing
values in your dataset. In WEKA, missing values are typically represented as empty
box.

SK BASHIRA 323103383L06
DATE: 09-02-2024

1. Filtering Missing Values: WEKA provides filters to handle missing values.


(i)ReplaceMissingValues filter, which replaces missing values with the mean,
median, or mode of the attribute values. You can find this filter in the "Filters"
tab under the "Unsupervised" category in the WEKA Explorer.

Before applying the ReplaceMissingValues filter we have 48 missing values for


the Attribute(water-projeact-cost-sharing).

After applying the ReplaceMissingValues filter we have 0 missing values for the
Attribute(water-projeact-cost-sharing) When we applied, this filter replaces
missing attribute values with calculation based on present values in the data set.

(ii)ReplaceMissingValuesUserConstant: This filter in WEKA is specifically


designed to handle missing attribute values within a dataset by replacing them
with a user-defined constant value.

SK BASHIRA 323103383L06
DATE: 09-02-2024

Before applying the ReplaceMissingValuesWithUserConstant filter the


Attributes have many missing values

After applying the ReplaceMissingValuesWithUserConstant filter we have


given value Y to get filled for the Attributes that have missing values

SK BASHIRA 323103383L06

You might also like