Social Media Ads Classification

SOCIAL MEDIA ads Classification

mini PROJECT

Students: Khadeeja Tabassum 135 Khadijah Abdul Azeem 136 Sumaiya Shoukath 151 Project Guide: Dr. B. V Ramana Murthy Project coordinator: Dr.Y.Shivani Yadao Team no. : 01

B.E CSE-C (3rd Year)

Machine learning is a field of artificial intelligence that focuses on the development of algorithms and models that can learn from and make predictions or decisions based on data, without being explicitly programmed. It involves training machines to recognize patterns, extract insights, and automate tasks, enabling them to improve their performance over time.

Domain: Machine Learning

INDEX

01. Introduction

05. Theoretical FW

11. Desision tree

06. Methodology

02. Tools & Technique

07. Classification

12.Development

03. Goals

13.conclusions

08. Training Model

04. Hypothesis

14.Bibiliography

09. Product Details

15.Acknowledgment

05. Problem statement

10.ClassificationReport

01. IntroduCTION

Social Media Ads Classification

The classification of social media ads is all about analyzing the ads for classifying whether your target audience will buy the product or not. It’s a great use case for data science in marketing.

• Classifying social media ads means analyzing your social media ads for finding the most profitable customers for your product who are more likely to buy the product.

Read more

02. Techniques & tools

MACHINE LEARNING TOOLS & TECHNIQUES

• A diverse set of algorithms and methodologies enable computers to learn from data and make predictions or decisions.
• These leverage statistical models, optimization algorithms, and data processing methods to extract patterns and insights, enabling automation and intelligent decision-making.

• Natural Language Processing (NLP).
• Text Classification Algorithms.
• Python
• Scikit- Learn
• Grid search cross validation

Read more

03. goals

Specific

General

• Improve ad targeting
• Optimize ad spending
• Enhance user experience
• Provide insights for marketers

• Develop a machine-learning model
• Preprocess ad data
• Evaluate model performance
• Optimize model hyperparameters

04.Hypothesis

Research Hypotheses

A diverse dataset will lead to high classification accuracy, effective preprocessing techniques will improve model performance, and optimized hyperparameters will enhance classification accuracy.

H1

H2

• Diverse dataset → High classification accuracy.
• Effective preprocessing → Improved model performance.
• Optimized hyperparameters → Enhanced classification accuracy.

H3

05.Problem statement

• Problem Statement: The problem statement of this project is to develop a machine learning model that can accurately classify social media ads into relevant categories.
• The challenge lies in effectively leveraging the textual content and attributes of the ads to enable targeted ad campaigns, optimize ad spending, and enhance user experience.
• By addressing this problem, we aim to provide marketers with a tool that can automate the ad classification process and improve the overall effectiveness of social media advertising strategies.

Read more

EXISTING MODEL

The existing model for social media ad classification relies on manual categorization and lacks automation, resulting in time-consuming and subjective processes. It struggles to handle large volumes of ads and may not capture the nuanced characteristics necessary for accurate classification.

Read more

06.Methodology

MACHINE learning model development

Data preprocessing

Clean and format the collected social media ad data, including tasks such as removing duplicates, handling missing values, and transforming the data into a suitable format for analysis.

Train and fine-tune a machine learning model using appropriate algorithms and techniques, leveraging feature extraction methods and optimization approaches to achieve accurate classification of social media ads.

Read more

07.classification

The dataset is downloaded from Kaggle. It contains data about a product’s social media advertising campaign. It contains features like:

• the age of the target audience
• the estimated salary of the target audience
• and whether the target audience has purchased the product or not

Read more

08.Training a Social Media Ads Classification Model

• Now let’s train a model to classify social media ads. First I’ll set the “Purchased” column in the dataset as the target variable and the other two columns as the features we need to train a model:
• x = np.array(data[["Age", "EstimatedSalary"]])
• y = np.array(data[["Purchased"]])

80%

ESCRIBE UN TÍTULO AQUÍ

Read more

09.product DETAILS DEVELOPMENT

10.classification report of the model

some of the important patterns in the dataset

• print(data.describe())
• print(data.isnull().sum())
• The first thing I want to explore is the ages of the people who responded to the social media ads and bought the product.

11.Decision tree

• split the data and train a social media ads classification model using the decision tree classifier:

import pandas as pd from sklearn.tree import DecisionTreeClassifier, export_graphviz import graphviz data = pd.read_csv('https://raw.githubusercontent.com/shivang98/Social-Network-ads-Boost/master/Social_Network_Ads.csv') X = data.iloc[:, :-1].values y = data.iloc[:, -1].values clf = DecisionTreeClassifier(max_depth=3) clf.fit(X, y) dot_data = export_graphviz(clf, out_file=None, feature_names=['Age', 'EstimatedSalary'], class_names=['Not Purchased', 'Purchased'], filled=True, rounded=True, special_characters=True) graph = graphviz.Source(dot_data) graph.render('social_network_ads_tree')

12.development

Population and sample

fieldwork

Fieldwork involves actively gathering data by conducting observations, interviews, surveys, or experiments in real-world settings.

• The population sample is developed by gathering a diverse collection of social media ads from various platforms,
• This curated sample aims to capture the key characteristics and trends present in the larger population of social media ads

Read more

Read more

13.Conclusions

• Successful development of accurate machine learning model for social media ad classification.
• Effective preprocessing techniques enhanced model performance.
• Hyperparameter optimization improved classification accuracy.
• Valuable insights provided for marketers to enhance ad targeting, optimize spending, and improve user experience.

Discussion

14. Bibliography

Bibliographic references

https://thecleverprogrammer.com/2021/06/15/social-media-ads-classification-with-machine-learning/ https://colab.research.google.com/drive/1zQ3aAwHLQwRWcg26XwSaFhOeRMmiUDJ6?usp=sharing

https://github.com/shivang98/Social-Network-ads-Boost/blob/master/Social_Network_Ads.csv

Figures and Tables

-Figure 1

-Figure 3

-Figure 5

-Figure 2

-Figure 4

-Figura 6

15. ACKNOWLEDGMENT

• We would like to express our heartfelt appreciation to Dr. Ramana Murthy for his invaluable guidance and support as our project guide. We are also grateful to Shivani Yadav for her coordination and assistance throughout the project. Their expertise and encouragement have been instrumental in the successful completion of this endeavor.

Dr.B.V Ramana Murthy Dr.Y.Shivani Yadao (Project Guide) (Project coordinator)

Thanks foryour attention

Any question?