Bayesian Inference –
Concept
Bayesian Inference is a method of updating probabilities based on
new evidence
Based on conditional probability
Uses prior belief and observed data
Helps refine predictions as more data becomes available
Widely used in data science for uncertainty handling
Bayes' Theorem
Formula
P(A|B) = [P(B|A) × P(A)] / P(B)
Prior: initial belief before observing
data
Likelihood: probability of data given
hypothesis
Posterior: updated probability after
evidence
Example: updating probability of an
event after new observation
Getting Data – Sources
Multiple Sources Common Sources Data Structure
Data can be collected from Common sources include files, Data may be structured or
multiple sources databases, and web unstructured
Data Quality First Step
Quality of data affects analysis Data collection is the first step in
results data science
Methods of Data Collection
01 02 03
Reading data from local files Accessing data through web Web scraping for extracting
APIs online data
04 05
Converting raw data into usable format Preparing data for analysis
Standard Input and Output
Used for reading input and displaying output
Enables communication between programs
Useful in data pipelines and scripting
stdin stdout
Standard input stream Standard output stream
Usage in Python
[Link] [Link] Line-by-Line Command-Line
Used to read input Used to print output Processing Programs
Allows processing data Common in command-
line by line line programs
import sys
for line in [Link]:
print([Link]())
File Handling Basics
Files are used to store data permanently
Python provides functions to read files
Files must be opened before reading
Common file types include text and CSV
Proper handling ensures efficient data usage
Reading Files in Python
File Functions
open() is used to access files
read() reads entire file
readline() reads one line
readlines() reads all lines into list
Example Code
with open("[Link]") as f:
lines = [Link]()