Picture through the Idogram
. Introduction
When you hear the word data science, you probably think about two words: programming and statistics. In fact, the condition of learning statistics often discourages people from getting a career -in -day career. This does not help that most of the data science job detail seems to be that you need a PhD in the data so that the role in this role can be achieved, when the reality is quite different.
In the majority of data science, especially in tech companies that focus on product development, you need to know Applied data. This includes the use of current statistical framework to solve business problems. This is different from academic statistics (think to calculate complex formulas by hand). Instead, you just need to understand what a concept means, how to calculate it using existing libraries, and how to be translated. Here is an example: In the most practical data science scenario, it is enough to understand what the PC of 0.03 means and how to use it to decide the business, rather than knowing how to calculate it by hand.
In this article, I will give you examples of how I use data in my data science job, as well as with the resources I used to acquire this knowledge.
. How do I use data in my data science job
!! Experience
Most tech companies (Google, Meta, Spatif) have a great culture of experiment. They strictly test the features before making changes.
When testing A/B, I need to know the concepts of data like:
- Statistics power to determine the sample size required for the experiment
- Level of importance, p values, and confidence breaks for decision -making
There are times when P -Valves do not tell the whole story, where you will need to learn more complicated forms of analysis, such as the difference of the difference (DID). However, these are the concepts that I raised on the job by reading articles, asking questions and conversation with senior colleagues. You cannot and may not learn and remember every concept required through courses or even a university degree. I recommend that you choose the basic concepts that you need to get through data science interview and need to learn the rest of the job.
!! Modeling
Building machine learning model requires data information. However, in my experience, it has been enough to learn and how to learn and create the theory behind these algorithms.
Of course, it does not apply to every industry. A data scientist working in a special sector such as predictions, biostatics, or aconetrics must have deep statistical knowledge related to his field.
However, in my experience, when working in products or tech companies, they are more focused on the business effects and interpretation of these models, rather than the hardness of mathematics.
!! Data analysis
I also spend a lot of time analyzing the data to understand how the user is interacting with the product, provides recommendations for how to improve this experience. It usually contains descriptive statistics, where I create concepts, distribute customer, and compare data distribution. Most of the data -related questions, such as “why customer retention falls in the last 3 months,” can be solved with easy concepts and does not require the use of sophisticated statistics methods.
In fact, if you know the difference between the meaning, the median, and the format and can develop ideas like histograms and box plots, you are already aware of this type of analysis. Rarely, you may need to use modern reactionary techniques or create a time series model. Once again, this is something I usually learn from senior peers, documents and online lessons.
. Three resources to learn data science stats
I have a computer science degree and I was not taught any data. All the knowledge of my statistics comes from the resources I have found online, and I have compiled a list of the most helpers.
- Introduction to Udacity for Statistics Is suggested for full initial individuals and covers descriptive stats, abnormal figures and possibilities
- State Quest It is helpful when you want to learn specific concepts. For example, if you want to learn how to work, you can get a 20 -minute lesson that is specific to the title of this channel.
- Statistical Learning on EDX There is another great course that you can audit for free. This way of learning teaches you to apply statistical concepts in Azar, which is mostly related to data science jobs
. Techways
Although the idea of learning data science can probably seem threatening, most data science jobs require you to know the data you apply, which has the ability to implement statistics concepts to solve business problems. In my experience, this knowledge can be easily acquired through online courses and does not require a master’s degree in data.
The resources listed in this article should be sufficient to get you through data science interview data. Beyond this, any knowledge can be obtained on this article by reading articles and papers on this article, working with the current framework in your organization, and learning from senior data scientists.
Natasa Selorj Is a self -educated data scientist with a passion for writing. Natasa writes on everything related to science, which is a real master of all data titles. You can contact with it Linked Or check it out Utube channel.