1857
51
1
15

🛠 Contribute: Found a typo? Or any other change in the description that you would like to see? Please consider sending us a pull request in the public repo of the challenge here.

## 🕵️ Introduction

Open up the black box of online advertising with this problem! What makes people click on ads? 🤔 This is a question that digital marketers have been asking themselves for years! An ad can fail or succeed for reasons that might have nothing to do with its content or relatability. But can we bring some predictability to it with the help of AI?

Given some crucial information on an online ad, can you predict whether it will be clicked or not?

Understand with code! Here is getting started code for you.😄

## 💾 Dataset

The dataset provided contains advertisements that were shown along side search results and whether or not they were clicked on. For each advertisement, 11 properties are given:

• Click - 1 if the add was clicked, 0 if it was not clicked
• impression - number of sessions in which the Ad was shown to the userID who issued the query
• url_hash - a hash of the url on which the ad was displayed
• AdID - an id uniquely identifying an ad
• advertiserID - an id uniquely identifying an advertiser
• Depth - number of ads displayed to a user in a session
• Position - order of an ad in a display
• query_id - an id uniquely identifying a query
• keyword_id - an id uniquely identifying a search keyword
• title_id - an id uniquely identifying the title of the ad
• description_id - an id uniquely identifying a description of the ad
• user_id - an id uniquely identifying a user

## 📁 Files

Following files are available in the resources section:

• train.csv - (40000 samples) This csv file contains the attributes describing an advertisement along with the binary value denoting whether or not the advertisement was clicked on.

• test.csv - (10000 samples) File that will be used for actual evaluation for the leaderboard score but does not have the binary value denoting whether or not the advertisement was clicked on.

## 🚀 Submission

• Prepare a CSV containing header as click and predicted value as digit 0 or 1 respectively denoting whether or not the corresponding ad will be clicked upon.
• Name of the above file should be submission.csv.
• Sample submission format available at sample_submission.csv in the resorces section.

Make your first submission here 🚀 !!

## 🖊 Evaluation Criteria

During evaluation F1 score will be used to test the efficiency of the model where,

$F1 = 2 * \frac{precision*recall}{precision+recall}$