Unit 2 Data Literacy Class 9 AI NOTES Important for Exam

Share with others

Unit 2 Data Literacy Class 9 AI NOTES

Unit 2 Data Literacy Class 9 AI NOTES

Introduction to Data Literacy

Data literacy to collect, analyze, and show data in different graphical ways that make sense. In other words we can say that Data literacy is the ability to understand, interpret and communicate with data.

Various steps are involved in working with data. These steps can be shown with the help of Data Pyramid.

Let us understand different parts of Data pyramid

Moving up from the bottom

● Data is available in a raw form which is not very useful.
● Data is processed to give us information.
● Information leads us to knowledge of how things are happening.
● Wisdom allows us to understand why things are happening in a particular way.

Let’s understand Data Pyramid with a simple Traffic Light example

Who is Data Literate?

Data Literate is a person who can interact with data to understand the world around them.

How data literate person can do shopping over the internet.

A data literate person can –

● Filter the category as per the requirement – If the budget is low, select the price filter as low to high

● Check the user ratings of the products

● Check for specific requirements in the product

Data Literacy Process Framework

The data literacy framework provides guidance on using data efficiently and with all levels of awareness.
Data literacy framework is an iterative process.

What are Data Security and Privacy? How are they related to AI?

Data privacy is concerned with the proper handling of sensitive data including personal data and other confidential data, such as certain financial data and intellectual property data.

Here are examples of two things which may compromise our data privacy.

Downloading unverified Mobile Application.
Accepting the “Terms of Service” without reading.

Why Data Privacy is important?

A data breach at a government agency can put top secret information in the hands of an enemy state.

A data breach at corporation can put proprietary data in the hands of competitor.

A breach at a hospital can put personal health information in the hands of those who might misuse it.

The following best practices can help you to ensure data privacy:

● Understanding what data, you have collected, how it is handled, and where it is stored.
● Necessary data required for a project should only be collected.
● User consent while data collection must be of utmost importance.

What is Data Security?

Data security is the practice of protecting digital information from unauthorized access, corruption, or theft throughout its entire lifecycle.

Why is it important?

The most possible reasons why data security is more important now are:

Cyber-attacks affect all the people
The fast-technological changes will boom cyber attacks

Best Practices for Cyber Security

Cyber security involves protecting computers, servers, mobile devices, electronic systems, networks, and data from harmful attacks.

Do’s

Use strong, unique passwords with a mix of characters for each account.
Activate Two-Factor Authentication (2FA) for added security.
Download software from trusted sources and scan files before opening.
Prioritize websites with “https://” for secure logins.
Keep your browser, OS, and antivirus updated regularly.
Adjust social media privacy settings for limited visibility to close contacts.
Connect only with trusted individuals online.
Use secure Wi-Fi networks.
Report online bullying to a trusted adult immediately.

Don’t ‘s

Avoid sharing personal info like real name or phone number.
Don’t send pictures to strangers or post them on social media.
Don’t open emails or attachments from unknown sources.
Ignore suspicious requests for personal info like bank account details.
Keep passwords and security questions private.
Don’t copy copyrighted software without permission.
Avoid cyberbullying or using offensive language online.

Acquiring Data, Processing, and Interpreting Data

Types of data

Artificial Intelligence is crucial, with data serving as its foundation. We come across different types of information every day. Some common types of data include

Textual Data (Qualitative Data)	Numeric Data (Quantitative Data)
It is made up of words and phrases	It is made up of numbers
It is used for Natural Language Processing (NLP)	It is used for Statistical Data
Search queries on the internet are an example of textual data Example: “Which is a good park nearby?”	Any measurements, readings, or values would count as numeric data Example: Cricket Score, Restaurant Bill

Numeric Data is further classified as:

Continuous data is numeric data that is continuous. E.g. height, weight, temperature, voltage. It can take any value including fractions. It’s measured, not counted. It is represented using histograms or line graphs

Discrete data is numeric data that contains only whole numbers and cannot be fractional E.g. the number of students in the class – it can only be a whole number, not in decimals. It’s counted. It is represented using bar charts.

Types of Data used in three domains of AI

1. Computer Vision(CV): Visual Data like images or videos

2. Natural Language Processing(NLP): Textual Data like document, pdf files

3. Statistical Data(SD): Numeric data like tables or excel sheet.

Q1. Pick and Choose from the following(Quantitative data or Qualitative data)

Temperature
Shoe size
Gender
Weight of a person
Favorite Colour

Ans. Temperature, shoe size and Weight of a person is Quantitative data(we can do mathematical operation like addition, subtraction etc.)

Gender and Favorite Colour is Qualitative data(we can not do mathematical operation like addition, subtraction etc.)

Data Acquisition/Acquiring Data

Data Acquisition refers to the procedure of gathering data. This involves searching for datasets suitable for training AI models. The process typically comprises three key steps:

1. Data Discovery: In this we search for new datasets.

2. Data Augmentation: Data augmentation means increasing the amount of data by adding copies of existing data with small changes.

3. Data Generation: Generating new data if data is not available.

Let we understand all the three steps given above in detail with example.

Step1: Data Discovery

Let’s say we want to collect data for making a CV model for a self-driving car

We will require pictures of roads and the objects on roads. We can search and download this data from the internet. This process is called data discovery

Step2: Data Augmentation

Data augmentation means increasing the amount of data by adding copies of existing data with small changes for example The image given does not change, but we can get data on the image by changing different parameters like color and brightness so we can add new data by slightly changing the existing data.

Step3: Data Generation

Data generation refers to generating or recording data using sensors for example recording temperature readings of a building is an example of data generation and recorded data is stored in a computer in a suitable form.

Sources of Data

Primary Data Sources — Some of the sources for primary data include surveys, interviews, experiments, etc. The data generated from the experiment is an example of primary data.

Secondary Data Sources—Secondary data collection obtains information from external sources, rather than generating it personally. Some sources for secondary data collection include:

Kaggle: It is an online community of data scientists where we can access different types of data.
Google dataset search: This is a toolbox by Google that can search for data by name.
UCI: It is a collection of databases, domain theories and data generators in collaboration with the university of Massachusetts.

Difference between Good data and Bad data

Good Data	Bad Data
Information is well structured	Information is scattered.
It is accurate	Contains a lot of incorrect values.
It is consistent	Contains missing and duplicate values.
It is clearly presented	It is poorly presented.
It contains information which is relevant to our requirement.	It contains information which is not relevant to our requirement.

Data acquisition from websites

The process of collecting data from websites using software is called Web Scraping. There are different tools that can help us to collect data from websites. Web Scraping is not illegal but acquiring data without permission is illegal.

NOTE: During data acquisition, we need to make sure that the data source allows data scraping.

Ethical concerns in data acquisition

1. Bias: During data acquisition we should avoid any preferences or partiality in data.

2. Consent: We should take necessary permission before collecting or using an individual’s data.

3. Transparency: We should explain our intentions about how we are going to use the collected data.

4. Anonymity: We should protect the identity of the person who is the source of data.

5. Accountability: We should take responsibility for our actions in case of misuse of data.

Features of Data and Data Preprocessing

Usability of Data

There are three primary factors determining the usability of data:

1. Structure of Data: It defines how data is stored.

Data stored in spreadsheet with the details of each product is considered as a good structure of data.

Data stored in a text document with no set of organizing rules is considered as bad structure of data.

2. Cleanliness: Clean data is free from duplicates, missing values, outliers, and other anomalies that may affect its reliability and usefulness for analysis.

3. Accuracy: Accuracy indicates how well the data matches real-world values, ensuring reliability. Accurate data closely reflects actual values without errors, enhancing the quality and trustworthiness of the dataset.

NOTE: Kaggle assigns a usability score to the data sets that are present on the website based on scores given by the users of that data

Features of Data

Data features are the characteristics or properties of the data. They describe each piece of information in a dataset. For example, in a table of student records, features could include things like the student’s name, age, or grade. These features help us to understand and analyze the data.

In AI models, we need two types of features: independent and dependent

Independent features are the input to the model—they’re the information we provide to make predictions.

Dependent features, on the other hand, are the outputs or results of the model—they’re what we’re trying to predict.

Data Processing and Data Interpretation

Data Processing Data processing helps computers understand raw data. Use of computers to perform different operations on data is included under data processing.

Data Interpretation It is the process of making sense out of data that has been processed. The interpretation of data helps us to answer critical questions using data.

keywords related to Data

1. Acquire Data– Acquiring data is to collect data from various data sources.

2. Data Processing– After collecting raw data, it is processed to derive meaningful information from it.

3. Data Analysis – Data analysis is to examine each component of the data in order to draw conclusions.

4. Data Interpretation – It is to be able to explain what these findings/conclusions mean in a given context.

5. Data Presentation– In this step, we select, organize, and group ideas and evidence in a logical way.

How to interpret Data?

There are two ways to interpret data-

1. Quantitative Data Interpretation It is made on numerical data. It helps us answer questions like “when,” “how many,” and “how often”, For example – (how many) numbers of likes on the Instagram post, number of visits on website, recording height of students in class.

Data Collection Methods for Quantitative Data Interpretation

1. Interviews: Quantitative interviews play a key role in collecting information.

2. Polls: A poll is a type of survey that asks simple questions to respondents. Polls are usually limited to one question.

3. Observations: Quantitative data can be collected through observations in a particular time period

4. Longitudinal Studies: A type of study conducted over a long time

5. Survey: Surveys can be conducted for a large number of people to collect quantitative data.

Steps to Quantitative Data Analysis

1. Relate measurement scales with variables

2. Connect descriptive statistics with data

3. Decide a measurement scale

4. Represent data in an appropriate format

2. Qualitative Data Interpretation tells us about the emotions and feelings of people. It is focused on insights and motivations of people. for example Reviews by customer.

Methods to collect data for– Qualitative Data Interpretation

Record keeping: This method uses existing reliable documents as the data source. It is similar to going to a library.

Observation: In this method, the participant’s behavior and emotions are observed carefully

Case Studies: In this method, data is collected from case studies.

Focus groups: In this method, data is collected from a group discussion on relevant topic.

Longitudinal Studies: This data collection method is performed on the same data source repeatedly over an extended period.

One-to-One Interviews: In this method, data is collected using a one-to-one interview.

Steps to Qualitative Data Analysis

1. Collect Data

2. Organize

3. Set a code to the Data Collected

4. Analyze your data

5. Reporting

Qualitative Data Interpretation vs Quantitative Data Interpretation

Qualitative Data Interpretation	Quantitative Data Interpretation
Provides insights into feelings and emotions	Provides insights into quantity
Answers how and why	Answers when, how many or how often
Methods – Interviews, Focus Groups	Methods – Assessment, Tests, Polls, Surveys
Example question – Why do students like attending online classes?	Example question-How many students like attending online classes?

Types of Data Interpretation

There are three ways in which data can be presented:

1. Textual Data Interpretation: The data is usually mentioned in the form of text, usually in a paragraph. It is used when the data is not large and can be easily comprehended by reading. Textual presentation is not suitable for large data.

2. Tabular Data Interpretation: Data is represented systematically in the form of rows and columns. ▪ Title of the Table refers the description of the table content. Column Headings contains the description of information contained in columns. for example

3. Graphical Data Interpretation: In this data is represented in the form of graphs for example

a. Bar Graphs: In a Bar Graph, data is represented using vertical and horizontal bars.

b. Pie Charts: This chart have the shape of a pie and each slice of the pie represents the portion of the entire pie allocated to each category. It is a circular chart divided into various sections (think of a cake cut into slices).

c. Line Graphs: A line graph is created by connecting various data points. It shows the change in quantity over time.

Importance of Data Interpretation

Informed Decision Making: A decision will be good when you understand or interpret the data clearly before making it. for example: School can custom design the chairs and tables for a class if the average height of students is known.

Reduced Cost: Cost can be reduced if we identify our needs. for example: Restaurant can drop some dishes which are not popular and got bad reviews.

Identifying needs: We can identify needs of people by data interpretation. for example: White sauce pasta is popular among children of age group 12 to 18 years.

Data Visualization Using Tableau

Download Tableau public with the help of an adult using this link – https://public.tableau.com/en-us/s/download

Install the package via the install wizard.

Once installed, double click the program to open the Tableau Public Desktop application.

Once open, this is what you should see.

Now click on the following icon (marked with arrow)

The following screen will appear.

Now click on the “Connect to Data” and the following screen will appear.

Before going further let we see the data which is already created in excel and saved with name “Sample” on Desktop or somewhere else.

To pull in the data, click on Microsoft Excel in the top left corner. Open dialog box appear on screen. Browse your file and click open.

A screen will appear on screen and then click on sheet as shown below

After clicking on Sheet1, another screen will appear

Click and drag “Name” up and to the right, releasing it next to the word Columns when a little orange arrow appears.

Now drag “Marks” to Rows

Tableau made a bar graph as shown below

Let’s explore another way of visualization

First, we’ll start by duplicating our current bar chart sheet. This will create an exact copy in a new sheet.

You’ll do this by right clicking “Sheet 1” and selecting “Duplicate”.

In the upper right corner, click “Show Me”.

We will see all of the different types of visualizations that Tableau can create.

Select “Packed Bubbles”.

Tableau quickly transformed our bar chart to a chart of bubbles.

Aman score the highest marks because it is the biggest circle.

We can make the text a little more fun and easier to read.

To do that, click the label square

This opens up a box that allows us to change the font and text size.

Let’s change the font size to 12 and the font to “Chalkboard”.

We have our complete bubble chart now!