Artificial Intelligence Unit 1 PDF

21CSC206T-Artificial Intelligence UNIT - 1 Introduction to AI 1  Topics Introduction to AI, AI techniques Problem - solving with AI AI Models Data acquisition and learning aspects in AI Problem solving-Problem solving process Form...

21CSC206T-Artificial Intelligence UNIT - 1 Introduction to AI 1  Topics Introduction to AI, AI techniques Problem - solving with AI AI Models Data acquisition and learning aspects in AI Problem solving-Problem solving process Formulating problems Problem types and characteristics Problem space and search Toy Problems–Tic-tac-toe problems Missionaries and Cannibals Problem Real World Problem–Travelling Salesman Problem 2 AGENDA Data acquisition and learning aspects in AI Knowledge discovery- Data Mining and Machine Learning COmputational Learning Theory (COLT) Neural and evolutionary computation Intelligent agent and multi-agent systems 3 Knowledge discovery- Data Mining and Machine Learning Data mining is actually the core step in Knowledge Discovery in Databases (KDD) process. Some preprocessing steps before data mining and post processing steps after data mining are to be completed to transform the raw data as useful knowledge. KDD is an iterative process that transforms raw data into useful information. 4 1. Data cleaning Noise and irrelevant data are removed from the large data set. This is a very important preprocessing step because the outcome would be dependent on the quality of selected data. Data cleaning is Removal of duplicate records Enter logically correct values for missing record Remove unnecessary data fields Standardize data format Update data in a timely manner etc. 5 2. Data transformation With the help of dimensionality reduction or transformation methods, the number of effective variables is reduced and only useful features are selected to depict data more efficiently based on the goal of the task. In short, data is transformed into appropriate form making it ready for data mining step. 3. Selection of data mining task Based on the objective of data mining, appropriate task is selected. Some common data mining tasks are classification, clustering, association rule discovery, sequential pattern discovery, regression and deviation detection. We can choose any of these tasks based on whether we need to predict information or describe information. 6 4. Selection of data mining algorithm Data mining is the actual search for patterns from the data available using the selected data mining method. Appropriate method(s) is to be selected for looking for patterns from the data. The model and parameters that might be appropriate for the method are to be decided. Some popular data mining methods are decision trees and rules, relational learning models, example based methods etc. 7 5. Pattern evaluation This is a post processing step in KDD which interprets mined patterns and relationships. If the pattern evaluated is not useful, then the process might again start from any of the previous steps, thus making KDD an iterative process. 7. Knowledge consolidation This is the final step in Knowledge Discovery in Databases (KDD). The knowledge discovered is consolidated and represented to the user in a simple and easy to understand format. Mostly, visualization techniques are being used to make users understand and interpret information. 8  The mining process includes data cleaning, preprocessing, identifying, interpreting the patterns, understanding the application and generating the target data It acts as a tool and holds core part in business Intelligence Machine learning Machine learning is a field concerned with study of algorithms that will improve its performance with experience, main focus is on improving performance of agent. 9  COmputational Learning Theory (COLT) By defining formal mathematical models, the efficiency and complexity in terms of computation, prediction and feasibility of algorithms are analyzed. Computation learning theory finds its importance in machine learning, pattern recognition and statistics. There are two frameworks for analysing the patterns 1. Probably Approximately Correct (PAC) It identifies the class of hypothesis that can/cannot be learnt 2. Mistake bound It tries to learn target functions to series of trails. 10  Neural and evolutionary computation This method enabled to speed up the mining of data Evolutionary computing is related to the study of biological properties like genetic algorithms. Telecom domain to financial decision making, with optimization as base criteria. In case of neural computing, neural behaviour of human brain is stimulated to enable machine to learn. ANN is configured for specific application like pattern recognition or classification problem. 11  Intelligent agent and multi-agent systems These type of intelligent system allows timely decision making in complex scenarios. An agent is simple software program that assists user. An intelligent agent is flexible in terms of its action to get the desired output. It is goal directed, reacts with environment and acts accordingly. Complex tasks and decision-making demand combination of more than one percept of different intelligent systems which can be done by Multi-Agent System (MAS). In MAS every agents capacity and its computation efficiency is exploited so that overall performance is improved. 12  Multi perspective integrated intelligence For problem solving each and every individual have their own perspective. Some extra information may/may not present in the different perspective. So based on that the decision making can be done. Exploiting and utilizing information from different perspectives to build up an intelligent system giving accurate results or helps in decision making called Multi Perspective Intelligence (MPI) frame work. Taking feedback. Collected information can be discrete or continuous in nature. This approach works in association with respect to the application it is chosen for. There is a fuzzy line of distinction between them. 13