![Setting Actions around the Goal
Post defining the problem statement using the 4W framework, actions can be set
using the DOIT principle as described below -
● Describe the specific pain-points or goals through the 4W canvas listing
down the stakeholders involved]
● Outline multiple different ways to reach the goal or solve each problem
● Identify the consequences of each of the above options outlined
● Take the most useful option and apply. Continue to check progress towards
the goal while thinking and working around the ethics](https://image.slidesharecdn.com/artificialintelligence-classix-240907085405-939b0919/85/Artificial-Intelligence-PPT-Class-IX-pdf-10-320.jpg)
Artificial Intelligence PPT- Class IX.pdf
- 2. Unit 1 - AI Reflection, Project Cycle and Ethics
- 3. Introduction to AI ● Artificial Intelligence refers to the ability of a machine or application to carry out a task that requires some degree of intelligence as compared to human intelligence. ● The term Artificial Intelligence was coined by John McCarthy. ● There are three domains of AI - Data and Data Science, Computer Vision, Natural Language Processing.
- 4. ● Data Science is a field that uses scientific methods, processes, algorithms and systems to extract knowledge and insights from many structured and unstructured data to apply in AI applications. ● Computer Vision refers to the training of computers to have a vision somewhat like humans so that machines can accurately identify and classify objects. ● Natural Language Processing (NLP) is a subfield of AI that revolves around the interactions between computers and human language in the form of input speech and output speech.
- 5. Advantages of AI 1. It has less room for errors 2. It exhibits right decision making 3. It can work 24 X 7 Disadvantages of AI 1. It is expensive to implement 2. AI depends on hardware and software efficacy 3. It can perform restrictive work only.
- 6. THE AI PROJECT CYCLE When an AI based project is undertaken, it goes through some phases from its initiation to closure which are collectively known as the AI Project Cycle. The various phases are - 1. Problem Scoping 2. Data Acquisition 3. Data Exploration 4. Modelling 5. Evaluation 6. Deployment
- 8. I. PROBLEM SCOPING This phase involves the following activities - a) Understanding the aim and scope of the project b) Understanding the expected outcomes of the project c) Understanding the stakeholders’ expectations d) Evaluating the probable success metric As the aim is to develop an AI based solution for a problem, choose a theme, choose a topic around the theme and then list the problems from the domain of the chosen theme.
- 9. 4W’s framework in Problem Scoping Setting goals for a project is merely insufficient. Rather the problem statement must be clearly defined using the 4Ws framework stating the WHO, WHAT, WHERE/WHEN and WHY of a problem. This helps in identifying the four crucial factors - 1. WHO are the stakeholders directly/indirectly getting affected by the problem 2. WHAT is the nature of the problem and is the evidence proving that is a problem requiring a solution 3. WHERE and WHEN did the problem arise. It also involves describing the context of the problem 4. WHY this problem should be solved and how it will benefit the stakeholders.
- 10. Setting Actions around the Goal Post defining the problem statement using the 4W framework, actions can be set using the DOIT principle as described below - ● Describe the specific pain-points or goals through the 4W canvas listing down the stakeholders involved] ● Outline multiple different ways to reach the goal or solve each problem ● Identify the consequences of each of the above options outlined ● Take the most useful option and apply. Continue to check progress towards the goal while thinking and working around the ethics
- 11. II. DATAACQUISITION Data Acquisition refers to the processes, methods or systems that are used to collect information related to a certain theme or objective, to document or analyse some phenomenon. Data Acquisition mainly involves - ● Grouping together relevant data features in a logically related structure ● Being clear about the relationship of data inside and outside the logical data structure ● Using consistent and standardised terminologies and format
- 12. Example - If you want to make an Artificially Intelligent system which can predict the salary of any employee based on his previous salaries, you would feed the data of his previous salaries into the machine. This is the data with which the machine can be trained. Now, once it is ready, it will predict his next salary efficiently. The previous salary data here is known as Training Data while the next salary prediction data set is known as the Testing Data. For better efficiency of an AI project, the Training data needs to be relevant and authentic. In the previous example, if the training data was not of the previous salaries but of his expenses, the machine would not have predicted his next salary correctly since the whole training went wrong. Similarly, if the previous salary data was not authentic, that is, it was not correct, then too the prediction could have gone wrong.
- 13. Finding Data Sources Data is the most crucial part of any AI project. To ensure quality of data, it must be acquired from reliable sources. Following can be categorised as reliable data sources - 1. Interview 2. Survey 3. Observation 4. Application Programming Interface (API) 5. Web Scraping (using bots to extract content and data from a website) 6. Sensors 7. Cameras 8. Problem Reports 9. Internet
- 14. Creating a System Map System maps are created to understand complex issues with multiple factors affecting each other. In a system every element is interconnected and has several chains of cause and effect. With the help of System Maps, one can easily define a relationship amongst different elements which come under a system LOOPY: a tool for thinking in systems (ncase.me) A System Map comprises of - ● Elements - These are different, discrete elements within the system ● Interconnections - These are the relationships that connect the elements.
- 15. Rules for creating a System Map - 1. Circle represents elements 2. Arrows are used to represent relationships 3. + and - signs are indicators of the nature of relationship ● Arrowhead depicts the direction of the effect and the sign (+ or -) shows their relationships ● If the arrow goes from X to Y with a + sign, it means that both are directly related to each other. That is, if X increases, Y also increases. ● If the arrow goes from X to Y with a - sign, it means that both are inversely related to each other. That is, if X increases, Y would decrease and vice versa
- 16. III. DATA EXPLORATION Data Exploration involves exploring data and looking for a pattern so as to choose the appropriate model for the project that can solve the problem. It also includes cleaning and normalization of data in order to standardise and correlate the data. An important aspect of data exploration is Data Visualization
- 17. Data Visualization Data Visualization refers to the process of representing data visually or graphically by using elements like charts, graphs, diagrams, maps etc. Its important to visualise data because - ● It explores data with presentable results ● It's easy to interpret and comprehend data ● Bulk data can be represented in a collective visual form. ● It is useful for combining categories of data and reducing the data for processing.
- 18. Data Visualisation Tools Some of the most commonly used data visualization tools are - 1. Scatter Chart 2. Bubble Chart 3. Line Graph 4. Pie Chart 5. Bar Graph 6. Histogram 7. Choropleth 8. Heat Map 9. Timeline 10. Node Link Diagram 11. Word Cloud
- 19. IV. MODELLING The process of developing an AI algorithm is called AI modelling where code is developed to produce intelligent outcomes using a set of data. AI modelling can be classified as follows -
- 20. Rule Based Approach - It refers to the Al modelling where the rules are defined by the developer. The machine follows the rules or instructions mentioned by the developer and performs its task accordingly.A drawback/feature for this approach is that the learning is static. The machine once trained, does not take into consideration any changes made in the original training dataset. Learning Based Approach - Refers to the AI modelling where the machine learns by itself. Under the Learning Based approach, the AI model gets trained on the data fed to it and then is able to design a model which is adaptive to the change in data.
- 21. V. EVALUATION Evaluation is the process of understanding the reliability of any AI model, based on outputs by feeding test dataset into the model and comparing with actual answers. The developed model is evaluated for accuracy and performance using new data so as to determine if the developed model is deployable or not. Few terms to be familiar with while evaluating an AI model - ❏ True Positive ❏ False Positive ❏ True Negative ❏ False Negative
- 23. VI. DEPLOYMENT Deployment as the final stage in the AI project cycle where the AI model or solution is implemented in a real-world scenario. Key Steps in Deployment Process the key steps involved in the deployment process: a. Testing and validation of the AI model b. Integration of the model with existing systems c. Monitoring and maintenance of the deployed model. Some examples of successful AI projects that have been deployed in various industries, such as self driving cars, medical diagnosis systems, and chatbots.
- 24. AI ETHICS
- 25. Ethics are a crucial part of an AI project and inculcate acceptable behaviours in all parties involved and ensure fair use and play of resources. Though ethics and morals have a slight difference -
- 26. AI Ethics Principles The following principles in AI Ethics affect the quality of AI solutions ▪ Human Rights ▪ Bias ▪ Privacy ▪ Inclusion
- 27. AI Bias Bias literally means inclination or prejudice for or against one person or group, especially in a way considered to be unfair. When AI programs, tools and algorithms exhibit any kind of bias, it is known as AI Bias. It often comes from the collected data and one should be careful about the following things - ➢ Does your data equally represent all the sections of the included populations? ➢ Will your AI learn to discriminate against certain groups of people? ➢ Does your AI exclude some people? ➢ What are some other biases that might appear in an AI?