Role of Data & Information

The names of two online surveying tools are:

1. Web forms (Google Form)
2. Emails

True

Reason — Information is the processed form of data.

False

Reason — Data available in the form of video is considered as video data, and video data is used in developing AI applications such as online lectures, videos captured through mobile phones, and YouTube videos.

True

Reason — X (previously known as Twitter) API helps in extracting tweets for data analysis purpose.

True

Reason — Platforms like Kaggle offer a huge number of datasets for different topics and provide datasets for machine learning purpose.

True

Reason — Video data contains images and audio together, therefore storage of video data takes more space in memory.

False

Reason — Data refers to the raw facts that are generated by almost all activities of our daily routine. A collection of some digits is raw data and does not have meaning or context, therefore it is not information.

True

Reason — To design an accurate AI system, relevant data must be identified, collected, and explored.

True

Reason — To learn patterns and relationships, AI models, such as deep learning neural networks, require extensive training with the help of data.

False

Reason — AI models can analyse medical imaging data, such as X-rays, CT scans, and MRIs, to help in disease detection and diagnosis in the healthcare industry.

True

Reason — Surveys can be done online or offline using web forms or emails. Users are asked a number of important questions, and their replies are recorded as text, audio, or video.

All of these

Reason — We are living in a digital world where electronic gadgets like mobile phones, sensors, and machines are generating and collecting a huge amount of data.

Processed data

Reason — Information is the processed form of data.

To train AI models

Reason — The primary purpose of data in AI is to train AI models. AI algorithms use huge amount of data to learn and make predictions.

Information

Reason — A student’s phone number is meaningful data related to a specific person and can be used for communication. Therefore, it is considered information.

Data

Reason — A list of student names of a class does not have meaning or context and therefore it is considered as data.

All of these

Reason — Data is represented in many forms such as text, image, audio, and video, and can be collected through newspapers, audio files, and images.

Natural Language Processing

Reason — Natural Language Processing would be suitable to process the data from an article. Large text data collections, such as books and articles, are utilised to train AI models to understand language and context.

All of these

Reason — Platforms such as Netflix, Amazon, Spotify, YouTube, etc., use AI-based recommendation systems to recommend personalised content to users.

Text

Reason — Data visualisation allows the graphical representation of data in the form of charts, graphs, or diagrams.

To get new and updated data

Reason — Data continuity means to get new and updated data.

1. Crowdsourcing is used to collect data from a large number of people.
2. Web scraping helps in getting useful information from websites and other online tools.
3. UCI Machine Learning Repository provides many datasets.
4. Mean and median are examples of descriptive statistics.

Data is important in AI-based medical imaging systems because AI models can analyse medical imaging data, such as X-rays, CT scans, and MRIs, to help in disease detection and diagnosis in the healthcare industry. By training on labelled medical images, AI algorithms can learn to recognise patterns associated with various conditions and help radiologists and physicians in making more precise decisions.

Some APIs help collect relevant data by allowing developers to get and use specific data from online platforms. APIs can be used to get data from social media sites, weather services, banking databases, and public data sources. For example, X (previously known as Twitter) API helps in extracting tweets for data analysis purpose.

Data can be collected through crowdsourcing by using crowdsourcing tools that let people from all over the world work together to acquire data. Tasks and instructions are given for collecting data, and people who are part of the crowd help in collecting data. This is a good method to collect a huge amount of data.

Data refers to the raw facts that are generated by almost all activities of our daily routine. Data is everything that we can get by reading, writing, speaking, and seeing. It can be made up of numbers, alphabets, symbols, or a mix of any or all of these elements. The collection of data or facts is considered as a dataset.

Various types of data are as follows:

1. Text Data: Data available in the form of text is considered as text data. Examples of text data are research articles, any information available on websites in text form, information available in books, newspapers, etc.
2. Image Data: Pictures are images that can be processed as image data. Examples of image data are the images related to a specific subject.
3. Audio Data: This type of data contains only audio files. Examples of audio data are audio clips of songs, recordings of speeches, recordings of lectures, news, etc.
4. Video Data: Data available in the form of video is considered as video data. Examples of video data are movies, online lectures, videos captured through mobile phones, YouTube videos, etc.

The various methods of collecting relevant data in AI project development are as follows:

1. Survey: Surveys can be done online or offline using web forms or emails. Users are asked a number of important questions, and their replies are recorded as text, audio, or video.
2. Observations: Data can be collected through observations by keeping track of people, things, or events in their natural environment and recording the observations.
3. Data Logging and Sensor Deployment: Sensors or data loggers are used to collect useful environmental data for AI projects that use the Internet of Things (IoT) or sensor-based data collection.
4. Web Scraping: Web scraping helps in getting useful information from websites and other online tools by crawling web pages and extracting structured or unstructured data.
5. Application Programming Interfaces (APIs): APIs allow developers to get and use specific data from online platforms such as social media sites, weather services, banking databases, and public data sources.
6. Collaborations and Data Partnerships: Data can be collected by making agreements to share data or by building partnerships with organisations that possess proprietary or specialised datasets.
7. Crowdsourcing: Crowdsourcing tools let people from all over the world work together to acquire data by performing assigned tasks and following given instructions.
8. Existing Datasets: Data can be collected from publicly available datasets provided by platforms such as Kaggle, UCI Machine Learning Repository, government data portals, and research institutions.

The various applications of data in AI are as follows:

1. Natural Language Processing (NLP): NLP applications rely extensively on data. Large text data collections, such as books, articles, and social media postings, are utilised to train AI models to understand language and context. This enables the AI system to generate accurate translations, understand sentiments, and answer queries based on the data’s knowledge.
2. Autonomous Vehicles: Data is essential for the development of autonomous vehicles. Sensors and cameras collect a huge amount of data about the surrounding environment, such as road conditions, other vehicles, and pedestrians. This data is used to train AI models to perceive and understand their surroundings, make decisions in real time, and drive safely.
3. Recommendations: Data plays a vital role in designing AI-based recommendation systems. Platforms such as Netflix, Amazon, Spotify, and YouTube use AI-based recommendation systems to recommend personalised content to users based on viewing history, purchase behaviour, and preferences.
4. Fraud Detection: AI models are used to analyse a huge amount of transactional data in fraud detection systems. By analysing data patterns and anomalies, these models help in identifying potentially fraudulent activities and generating alerts.
5. Medical Imaging: AI models can analyse medical imaging data, such as X-rays, CT scans, and MRIs, to help in disease detection and diagnosis in the healthcare industry. By training on labelled medical images, AI algorithms learn to recognise patterns associated with various conditions.

To collect data on classmate's career choices, the steps of identifying and collecting relevant data are as follows:

Identifying Relevant Data

1. Defining Data Requirement: The first step is to define the data requirements by understanding the problem to be solved. In this case, data related to classmates’ career choices is required.
2. Acquiring Domain Expertise: Discussion with teachers or career counsellors who have in-depth understanding of career options can help in identifying relevant data and deciding what information is important.
3. Reviewing Literature: Reviewing articles, reports, or studies related to career choices can help in understanding what type of data should be collected.

Collecting Relevant Data

1. Survey: A survey can be conducted by asking classmates questions through questionnaires, online forms, or interviews. Their responses can then be recorded and used as data.

The most suitable AI technique for identifying a disease from X-ray images is Machine Learning used with image recognition.

This technique is suitable because:

• X-ray images are a type of unstructured data, which means they cannot be analysed using simple rules or formulas.
• Machine Learning allows a computer to learn patterns from examples. By showing the AI many X-ray images of healthy patients and diseased patients, it learns how they look different.
• The AI system compares shapes, shadows, and patterns in the X-ray images to find signs of disease.
• Once trained, the AI can identify whether a new X-ray image shows the disease or not.