Naïve Bayes Classifier Algorithm common technique to solve such classification problems. For e.g.

, if we have two classes

with multiple features and need to separate them efficiently. When we classify them
o Naïve Bayes algorithm is a supervised learning algorithm, which is based
using a single feature, then it may show overlapping. Define Pattern Recognition? Describe about Design principles of pattern
on Bayes theorem and used for solving classification problems.
Define MYCIN and EMYCIN? Describe about Knowledge recognition system and Statistical Pattern recognition?
o It is mainly used in text classification that includes a high-dimensional Representation in expert systems Pattern Recognition System Pattern is everything around in this digital world. A
training dataset. MYCIN is a computer-based consultation system designed to assist physicians in the pattern can either be seen physically or it can be observed mathematically by
applying algorithms. In Pattern Recognition, pattern is comprises of the following
o Naïve Bayes Classifier is one of the simple and most effective Classification diagnosis of and therapy selection for patients with bacterial infections. In addition to
two fundamental things:
algorithms which helps in building the fast machine learning models that the consultation system itself, MYCIN contains an explanation system which can answer
can make quick predictions. simple English questions in order to justify its advice or educate the user. The system's  Collection of observations

o It is a probabilistic classifier, which means it predicts on the basis of

knowledge is encoded in the form of some 350 production rules which embody the  The concept behind the observation
clinical decision criteria of infectious disease experts. Much of MYCIN's power derives
the probability of an object. from the modular, highly stylized nature of these decision rules, enabling the system to  Differentiate between good and bad features.
o Some popular examples of Naïve Bayes Algorithm are spam filtration, dissect its own reasoning and allowing easy modification of the knowledge base.
 Feature properties
Sentimental analysis, and classifying articles. EMYCIN : An expert system shell implemented in INTERLISP <103>. EMYCIN is a
here are several basic principles and design considerations that are important in
Bayes' Theorem: domainin-dependent version of MYCIN, a production rule system <192> designed pattern recognition:
o Bayes' theorem is also known as Bayes' Rule or Bayes' law, which is used for medical consultations. Problem-specific knowledge is represented as production 1. Feature representation: The way in which the data is
rules where the antecedent is effectively a boolean function of predicates of attribute- represented or encoded is critical for the success of a pattern
to determine the probability of a hypothesis with prior knowledge. It recognition system. It is important to choose features that are
object-value triples and both the condition and action have a certainty value
depends on the conditional probability. relevant to the problem at hand and that capture the underlying
associated with them. Uses a backward chaining control strategy. Incorporates a
o The formula for Bayes' theorem is given as: sophisticated front-end to handle user interactions and facilities for explaining how
structure of the data.
2. Similarity measure: A similarity measure is used to compare the
conclusions were reached and answering questions. similarity between two data points. Different similarity measures
Where, What is knowledge representation? may be appropriate for different types of data and for different
P(A|B) is Posterior probability: Probability of hypothesis A on the observed event B. Humans are best at understanding, reasoning, and interpreting knowledge. Human problems.
P(B|A) is Likelihood probability: Probability of the evidence given that the probability knows things, which is knowledge and as per their knowledge they perform various 3. Model selection: There are many different types of models that
of a hypothesis is true. actions in the real world. But how machines do all these things comes under can be used for pattern recognition, including linear models,
P(A) is Prior Probability: Probability of hypothesis before observing the evidence. knowledge representation and reasoning. Hence we can describe Knowledge nonlinear models, and probabilistic models. It is important to
representation as following: choose a model that is appropriate for the data and the problem
P(B) is Marginal Probability: Probability of Evidence.
at hand.
Types of Naïve Bayes Model: o Knowledge representation and reasoning (KR, KRR) is the part of Artificial 4. Evaluation: It is important to evaluate the performance of a
There are three types of Naive Bayes Model, which are given below: intelligence which concerned with AI agents thinking and how thinking pattern recognition system using appropriate metrics and
o Gaussian: The Gaussian model assumes that features follow a normal contributes to intelligent behavior of agents. datasets. This allows us to compare the performance of different
algorithms and models and to choose the best one for the
distribution. This means if predictors take continuous values instead of o It is responsible for representing information about the real world so that problem at hand.
discrete, then the model assumes that these values are sampled from the a computer can understand and can utilize this knowledge to solve the 5. Preprocessing: Preprocessing is the process of preparing the
Gaussian distribution. complex real world problems such as diagnosis a medical condition or data for analysis. This may involve cleaning the data, scaling the
o Multinomial: The Multinomial Naïve Bayes classifier is used when the communicating with humans in natural language.
data, or transforming the data in some way to make it more
suitable for analysis.
data is multinomial distributed. It is primarily used for document o It is also a way which describes how we can represent knowledge in 6. Feature selection: Feature selection is the process of selecting a
classification problems, it means a particular document belongs to which artificial intelligence. Knowledge representation is not just storing data subset of the most relevant features from the data. This can help
category such as Sports, Politics, education, etc. to improve the performance of the pattern recognition system
into some database, but it also enables an intelligent machine to learn
The classifier uses the frequency of words for the predictors. and to reduce the complexity of the model.
from that knowledge and experiences so that it can behave intelligently
o Bernoulli: The Bernoulli classifier works similar to the Multinomial like a human.
classifier, but the predictor variables are the independent Booleans
variables. Such as if a particular word is present or not in a document. This
model is also famous for document classification tasks.

Describe about Principle Component Analysis (PCA) and Linear

Discriminant Analysis (LDA)?

Principal Component Analysis

Principal Component Analysis is an unsupervised learning algorithm that is used for the
dimensionality reduction in machine learning. It is a statistical process that converts the
observations of correlated features into a set of linearly uncorrelated features with the
help of orthogonal transformation. These new transformed features are called
the Principal Components. It is one of the popular tools that is used for exploratory
data analysis and predictive modeling. It is a technique to draw strong patterns from
the given dataset by reducing the variances.
What is Linear Discriminant Analysis (LDA)?
Although the logistic regression algorithm is limited to only two-class, linear
Discriminant analysis is applicable for more than two classes of classification problems.
Linear Discriminant analysis is one of the most popular dimensionality reduction
techniques used for supervised classification problems in machine learning . It is
also considered a pre-processing step for modeling differences in ML and applications
of pattern classification.
Whenever there is a requirement to separate two or more classes having multiple
features efficiently, the Linear Discriminant Analysis model is considered the most