STATISTICS WITH

DATA SCIENCE
SIDDHARTHA DAS | CLASS - X - C | ROLL NO. - 13
DATA SCIENCE PORTFOLIO
 CONTENTS
01 APPLICATIONS OF DATA SCIENCE

HOW TO MAINTAIN PROPER STATISTICS IN

03
USE OF DIFFERENT FORMULAS IN DATA
SCIENCE
APPLICATIONS OF DATA SCIENCE
Business and Marketing: Data science is used to analyze consumer behavior, predict market trends, and
optimize marketing strategies. It helps businesses make data-driven decisions, target the right audience, and
personalize customer experiences.
Healthcare: Data science plays a crucial role in healthcare for tasks such as disease diagnosis, drug discovery,
treatment optimization, and patient monitoring. It enables the analysis of large medical datasets to identify
patterns and insights for better healthcare outcomes.
Finance and Banking: Data science is used in fraud detection, credit scoring, risk assessment, algorithmic
trading, and portfolio optimization. It helps financial institutions make informed decisions, detect anomalies, and
improve customer experience through personalized financial services.
Transportation and Logistics: Data science helps optimize transportation routes, reduce costs, and improve
logistics efficiency. It enables predictive maintenance of vehicles, demand forecasting, and real-time tracking for
efficient supply chain management.
Social Media and Sentiment Analysis: Data science techniques are used to analyze social media data, identify
trends, and understand customer sentiment. This information helps businesses gauge public opinion, develop
targeted marketing campaigns, and improve customer engagement.
Natural Language Processing (NLP): NLP, a subfield of data science, is used for tasks such as text classification,
sentiment analysis, machine translation, and chatbots. It enables computers to understand and process human
language, leading to applications in customer support, content generation, and information extraction.
HOW TO MAINTAIN PROPER STATISTICS IN DATA SCIENCE
Data Cleaning and Preprocessing: Start by thoroughly cleaning and preprocessing the data. This involves handling missing values,
removing outliers, and addressing inconsistencies. It is essential to ensure that the data is accurate and representative of the problem you
are trying to solve.
Descriptive Statistics: Calculate and analyze descriptive statistics to gain insights into the data. This includes measures such as mean,
median, standard deviation, and quartiles. Descriptive statistics provide a summary of the data distribution and help identify patterns and
outliers.
Data Visualization: Utilize data visualization techniques to present the data in a visually appealing and understandable manner. Use
histograms, scatter plots, box plots, and other visualizations to explore relationships, identify trends, and communicate findings effectively.
Statistical Inference: Apply statistical inference techniques to draw conclusions and make predictions from the data. This involves
hypothesis testing, confidence intervals, and regression analysis. Statistical inference helps in determining the significance of relationships,
validating models, and making predictions based on the data.
Experimental Design: If conducting experiments or A/B testing, design experiments carefully to ensure unbiased and statistically valid
results. Consider factors like sample size, randomization, control groups, and statistical power. Proper experimental design helps in drawing
meaningful conclusions and avoiding spurious correlations.
Model Evaluation: Evaluate the performance of predictive models using appropriate statistical metrics. Common metrics include accuracy,
precision, recall, F1-score, and ROC curves. Model evaluation helps assess how well the model fits the data and its predictive capabilities.
Statistical Software and Tools: Utilize statistical software and tools such as R, and Python with libraries like NumPy, Pandas, and SciPy, or
dedicated statistical packages like SPSS or SAS. These tools provide a range of statistical functions and algorithms to facilitate data analysis.
Documentation: Maintain proper documentation of the statistical analysis performed, including the steps taken, assumptions made, and
results obtained. This documentation ensures transparency, and reproducibility, and allows for effective collaboration and sharing of
findings.
USE OF DIFFERENT FORMULAS
IN DATA SCIENCE
Descriptive Statistics
a. Mean: The average value of a set of numbers. Formula:
(Sum of all values) / (Number of values) Probability and Statistics
b. Median: The middle value in a sorted set of numbers. a. Probability: The likelihood of an event occurring.
Formula: Middle value or an average of two middle values Formula: (Number of favorable outcomes) / (Total number
c. Standard Deviation: A measure of the spread of data of possible outcomes)
around the mean. Formula: sqrt((Sum of squared b. Bayes' Theorem: A formula to update probability
differences from the mean) / (Number of values - 1)) estimates based on new evidence. Formula: P(A|B) =
d. Correlation: A measure of the relationship between two (P(B|A) * P(A)) / P(B)
variables. Formula: (Covariance of X and Y) / (Standard c. Central Limit Theorem: A theorem stating that the
Deviation of X * Standard Deviation of Y) sampling distribution of the mean tends to be normal,
Linear Regression regardless of the shape of the population distribution.
a. Simple Linear Regression: A formula to model the relationship Probability Distributions
between a dependent variable and an independent variable. a. Normal Distribution: A continuous probability distribution used
Formula: y = mx + b, where y is the dependent variable, x is the to model various natural phenomena. Formula: f(x) = (1/√(2πσ2))
independent variable, m is the slope, and b is the intercept. (e[-(x-μ)^2]/2σ^2)
b. Multiple Linear Regression: A formula to model the b. Poisson Distribution: A discrete probability distribution used to
relationship between a dependent variable and multiple model the number of events occurring within a fixed interval of
independent variables. Formula: y = b0 + b1x1 + b2x2 + ... + bnxn, time or space. Formula: P(x; λ) = (e^(-λ) * λ^x) / x!, where λ is the
where y is the dependent variable, x1, x2, ... xn are the average rate of events and x is the number of events.
independent variables, and b0, b1, b2, ... bn are the coefficients.
THANK
YOU