SMS Spam Classification Test phase, srs, design phase and code final deliverable

SMS Spam Classification Test phase, srs, design phase and code final deliverable

Project Domain / Category
Data Science/Machine Learning

Abstract / Introduction
As we recognise that, SMS (Short Message Service) is one of the in your price range and popular services and most used carrier in mobile network. It has excessive reaction rate and having right confidentiality with trusted and private service.

Due to that undesirable SMS called junk mail SMS will rise up so that it will generate different hassle to cellular user. To become aware of such junk mail message is one of the crucial challenges in internet and wi-fi network.

In this undertaking we are able to makes use of PYTHON textual content class approach to identify or classify unsolicited mail message.
We will locate accuracy, time and error rate with the aid of applying appropriate algorithms (together with NaiveBayes, NaiveBayesMultinomial and J48 and many others.) on SMS Dataset and we will additionally compare which algorithm is satisfactory for textual content classification.

Collect Data Set.
• Gathering the data for SMS unsolicited mail contains spam and non-spam messages.

Functional Requirements:

1. Pre-processing
• As most of the facts within the actual global are incomplete containing noisy and missing values. Therefore we need to follow Pre-processing for your records.

2. Feature Selection
• After the pre-processing step, we apply the characteristic selection algorithm, the algorithm which installation here is Best First Feature Selection algorithm.

Three. Apply Spam Filter Algorithms.
• Handle Data: Load the dataset and cut up it into training and take a look at datasets.
• Summarize Data: summarize the homes within the schooling dataset so that we will calculate chances and make predictions.
• Make a Prediction: Use the summaries of the dataset to generate a unmarried prediction.
• Make Predictions: Generate predictions given a test dataset and a summarized training dataset.
• Evaluate Accuracy: Evaluate the accuracy of predictions made for a test dataset as the percentage accurate out of all predictions made.
Four. Train & Test Data
• Split facts into seventy five% training & 25% trying out information units.

Five. Confusion Matrix
• Create a confusion matrix table to explain the overall performance of a classification model.

6. Accuracy
• Find Accuracy of all set of rules and evaluate.

Tools:
Python
Anaconda

Prerequisite:
Artificial intelligence Concepts, Machine studying.


Supervisor:
Name: Muhammad Tayyab Waqar
Email ID: tayyab.Waqar@vu.Edu.Pk
Skype ID: maliktayyab786_1