Thank you ❤. I'm not a techy person. However the way you handle the material is so smooth, so compelling that my fears have disappeared and I'm ready to start learning

Subscribe to our YouTube channel for data science tutorials and tips. Drop your email ID in the comments to get a free Data Science Interview Kit! Do it soon! Don't miss out!

1:37:00

bro what about data scientist in india ??

This Course is made more towards for what the real and practical advice for work that needs to be done in one video, not for actual learning of ML, DL, LLM, CV or really anything specific as a skill not exactly what I was Looking for.

Your voice and way of teaching ❤❤

thank you

I lost my recent started data science course

Day 1; 32:10

A hearty thanks,
❤
I just completed my 10th boards,
And was in search of a best career option for self with all the clarifications ,
And you this video, made me get it perfectly with no doubt remained in me ,

And really the ones who are contributed in makeing this video are psychologists, who actually knows how we (students) do think !

Really meant alot ❤❤❤

Wonderful voice. Quality information

Amazing course! I'm having a turn in my career after 10 years of school teaching, and this course is just what I needed to complement my studies in DS. Just an errata : at 2:13:20 you mentioned "data de novo" as your personal expression for the concept of "new data". I don't know you if it meant for this expression to be extracted from portuguese language or latin, but if you took it from portuguese, the adequate expression should be "Nova Data" or "Novo Dado". In portuguese "de novo" means the same as "again" or to do something one more time.

For my use as bookmark 4:52:00

This video is higly valuable. unfortunately this video is old and there is a high probability that much things has changed.

dude, you sound exactly like 3Blue1Brown. Are you two the same??

I m in class 11 can i learn this data science 🥲

This is really good, very well put together, concise and easy to understand. I am totally new to data science and this has seriously got me considering learning about it further. Hopefully with a view of working in this field one day. Thank you

masterpiece … thank uhh

Very good content !

🎯 Key Takeaways for quick navigation:

02:48 🌐 Data Science is inclusive analysis, involving all data to provide the most insightful answers to research questions.
14:33 🌐 Data Science involves diverse skills and backgrounds, encompassing coding, statistics, math, and domain expertise, making it a compelling career alternative.
23:09 🦄 Data Science Diversity: Data science is diverse, involving people with different goals, skills, and experiences working in various contexts, making it a rich and interconnected field.
24:34 🤖 Unicorn Analogy: The term "unicorn" is used in data science to describe a mythical data scientist with universal abilities in coding, statistics, design, business, and management. However, in reality, such individuals are rare, and collaboration among specialists is more common.
35:25 📊 Data Science vs. Statistics: Data science and statistics share common procedures, but they differ in training backgrounds, goals, and contexts, highlighting their conceptual distinctions despite overlapping elements.
47:08 🧠 Data science analyses are limited simplifications; humans are essential for interpretation and application. Overconfidence in algorithmic results can lead to incorrect conclusions.
48:25 🌐 Data science projects can't be neutral; algorithms reflect the biases of their creators. Good judgment is vital for the quality and success of a data science project.
58:25 🧮 Math is foundational in data science; understanding procedures, addressing issues, and some manual calculations are essential for informed decisions.
01:10:01 📈 When conducting data analysis, focus on maximizing the story to maximize value. Clearly align the narrative with specific goals, especially when answering client queries.
01:13:47 📊 In presenting data analysis, adhere to the principle of being minimally sufficient. Embrace simplicity, as emphasized by quotes like "Everything should be made as simple as possible, but not simpler."
01:19:33 🎯 Data science is goal-focused. When communicating results, provide specific, justifiable next steps based on the analysis. Consider the social, political, and economic context for actionable insights.
[01:34:28 URL](https://youtu.be/ua-CiDNNj30?t=5668s) 📘 Use narrative methods like Jupyter Notebooks or RMarkdown to document and share the data analysis process, allowing for transparency and understanding of conclusions.
[01:36:46 URL](https://youtu.be/ua-CiDNNj30?t=5806s) 🚀 Next steps after the tutorial: Explore coding in R or Python, dive into data visualization, brush up on statistics and math, explore machine learning, and consider community involvement in data science conferences and projects.
[01:39:27 URL](https://youtu.be/ua-CiDNNj30?t=5967s) 🌐 Data science is democratic and essential for everyone. Encourages learning to work with data intelligently and sensitively, emphasizing its fundamental importance for all.
02:12:52 🕵️‍♀️ Data scraping from webpages, PDFs, images, and media when no API is available. Code scraping examples in R and Python. Emphasizes respect for copyright and privacy to avoid legal issues.
[02:16:06 URL](https://youtu.be/ua-CiDNNj30?t=9766s) 🎙️ Interviews are valuable for new situations or audiences, offering open-ended information. Structured interviews have predetermined questions, while unstructured interviews are more conversational and varied.
[02:18:26 URL](https://youtu.be/ua-CiDNNj30?t=11006s) 📊 Surveys are effective for obtaining data quickly, but clarity in question wording and response scales is crucial. Beware of bias, and ensure the questions align with the audience's understanding.
[02:25:42 URL](https://youtu.be/ua-CiDNNj30?t=15442s) ⚗️ Laboratory experiments are crucial for determining cause and effect, requiring specialized training. They offer reliable information but can be time-consuming and expensive.
02:40:02 🧹 Spreadsheets excel in tasks like data browsing, sorting, rearranging, finding/replacing, formatting, transposing, tracking changes, creating pivot tables, and arranging output for consumption.
02:41:51 🔍 When working with spreadsheets, maintaining "Tidy Data" is crucial for easy transfer between programs. Tidy Data involves having equivalent columns for variables and rows for cases, ensuring one sheet per file, and maintaining a consistent level of measurement per file.
02:54:39 📊 SPSS (Statistical Package for the Social Sciences) is a desktop program used in academic and business research, known for its point-and-click interface and drop-down menus. The program is available for free for students, with paid versions for others.
02:59:14 📊 SPSS provides various options for data analysis, including descriptive statistics and visualization tools like stem-and-leaf plots and box plots.
03:02:22 🆕 JASP, a free and open-source alternative to SPSS, offers intuitive features, replicability, and includes Bayesian approaches.
03:09:09 💻 A wide range of data science tools, including SAS, Stata, MATLAB, Wolfram Alpha, RapidMiner, KNIME, SOFA Statistics, and more, are discussed with considerations for functionality, ease of use, community support, and cost.
03:34:45 🐍 Python, a general-purpose language, is popular in data science. It has a vast community, multiple versions (2.x and 3.x), and interfaces like Jupyter. Python's strength lies in numerous packages, including NumPy, Pandas, and scikit-learn.
03:40:12 💽 SQL, the language of databases, is crucial in data science. It excels in relational databases (RDBMS), with popular choices like Oracle, SQL Server, MySQL, and PostgreSQL. SQL minimizes data redundancy and is often used via GUIs like SQL Developer.
03:43:44 📊 SQL commands: Learn essential SQL commands, including SELECT, FROM, WHERE, and ORDER BY, for efficient data extraction and organization from relational databases.
03:44:41 ⚙️ Data Science Languages: C, C++, and Java serve as foundational languages in data science, particularly for the back end, offering speed and reliability.
04:06:51 🧮 Algebra is crucial to data science, enabling the combination of scores and various manipulations. Linear algebra, also known as matrix algebra, is the next step, representing data with vectors and matrices, making computations more efficient.
04:29:59 📈 Lowering the cost by 20%, from $500 to $400 per year, can increase sales by 33%, leading to a 7% increase in total revenue.
04:35:51 📊 Understanding Big O helps optimize algorithms, considering time and space complexity, crucial for efficient data processing.
04:41:19 🎲 Probability calculations involve adding or multiplying probabilities, considering overlaps and conditional probabilities.
04:49:31 🔄 Positive test result doesn't guarantee disease; Bayes theorem crucial for accurate probability calculations.
04:51:23 🎯 Focus on goals in data science; understand procedures, diagnose problems, and prioritize meaning.
05:09:40 📊 Tukey's ladder of powers helps transform skewed data; exploring numerical distributions aids in understanding data stability, outliers, and skewness.
05:11:04 📈 Descriptive statistics involve center, spread, and shape. Common measures include mode, median, mean for center; range, interquartile range, variance, and standard deviation for spread.
05:19:02 🔄 Measures of spread (range, interquartile range, variance, and standard deviation) have pros and cons; variance and standard deviation are less intuitive but more useful in data science.
05:20:47 📊 Understanding the shape of the distribution (symmetrical, skewed, unimodal, bimodal, uniform, u-shaped) is crucial for interpreting numerical summaries like mean and standard deviation.
05:26:01 📊 Inferential statistics involve sampling data from populations, adjusting for sampling error; common approaches include hypothesis testing (null and alternative hypotheses) and estimation.
05:51:10 🔄 Multicollinearity, the association between predictors, poses challenges in regression analysis. Methods like stepwise regression, commonality analysis, dominance analysis, and relative importance weights help address multicollinearity issues.
05:51:40 📊 Non-Normality, Non-Linearity, Multicollinearity, and Missing Data are common problems in modeling. Skewed distributions, outliers, and mixed distributions impact the assumptions of statistical procedures. Strategies include data transformation, polynomial terms, and reducing variables.
05:52:38 📉 Model Validation is crucial for assessing the generalizability of statistical models. Approaches like Bayesian methods, replication, holdout validation, and cross-validation help evaluate model performance on different data sets.
05:53:36 🛠️ Adopt a DIY (Do It Yourself) attitude in data science. Emphasize the importance of getting started, align methods with goals, focus on usability, and beware of trolls and critics. Acknowledge that no analysis is perfect, but the goal is to add value to the understanding of the data.

Made with HARPA AI

1:00:00

17:06