Monday, November 28, 2022

What Is Domain Knowledge In Data Science

Don't Miss

What Is Domain Knowledge

Why Domain Knowledge Is Important In Data Science?

In data science, the term domain knowledge is used to refer to the general background knowledge of the field or environment to which the methods of data science are being applied. Data science, as a discipline, can be thought of as the study of tools used to model data, generate insights from data, and make decisions based on data. They are generic tools applicable to many fields such as engineering, laws, medicine, finance, etc.

Incorporating Domain Knowledge Into Your Architecture And Your Model Can Make It A Lot Easier To Explain The Results Both To Yourself And To An Outside Viewer Every Bit Of Domain Knowledge Can Serve As A Stepping Stone Through The Black Box Of A Machine Learning Model

Developing machine learning models involves a lot of steps. Whether youre working with labeled or unlabeled data, you might think numbers are just numbers and that it doesnt matter what each of the features of a data set signifies when it comes to spitting out insights with the potential for true impact. Its true that there are tons of great machine learning libraries out there such as scikit-learn that make it straightforward to gather up some data and plop them into a cookie-cutter model. Pretty quickly, you might start to think theres no problem you cant solve with machine learning.

Frankly, thats a beginners mindset. You are not yet aware of everything you dont know. Data sets given in machine learning courses or the free ones you find online often have been groomed already and are convenient to use when applying machine learning models, but once you take your skills and knowledge out of the playpen and into the real world, youll face some additional challenges.

Lots of people believe that domain knowledge, or additional knowledge regarding the industry or area the data pertains to, is superfluous. And its kind of true. Do you need domain knowledge in the area youre developing the model? No. You can still produce fairly accurate models without it. Theoretically, deep and machine learning are black-box approaches. This means you can put labeled data into a model without deep knowledge of the area and without even looking at the data very closely.

Aim: What Sort Of Challenges A University Faces While Instrumenting A Data Science And Analytics Course

Professor Ramanathan: Relevance is a challenge that universities and institutions face on a real-time basis. Industry demands and requirements are constantly changing, and it is crucial to make sure the course content and structure address the changing professional environment. It becomes even more important since most data science students are professionals with relevant industry experience. Therefore, when teaching working professionals, we have to be cognizant of their work timing and ensure the content is compact and easily comprehensible.

Also Check: How To Create Website After Buying Domain

The Elephant In The Room

This situation is well illustrated by the famous elephant parable. Several blind persons, who have never encountered an elephant are asked to touch one and describe it. The descriptions are all good descriptions given the experience of each person, but they are all far from the actual truth because each person was missing data.

Picture credit: Saeed Mubarak

This problem could have been avoided with more data or with some contextual information derived from existing elephantine descriptions. Moreover, the effort might be better guided if it is clear what the description will be used for.

Case Study: Predicting Credit Card Delinquency

Data Science Domains  Stephane Andre

In this section, we will look at a case study that illustrates the importance of domain knowledge. Predicting credit card delinquency is a common problem in consumer finance, where a credit card provider must decide whether to issue credit cards to a particular customer. It also helps the provider make risk assessments and strategic decisions.

We will look at a small data science project that aims to predict delinquency in credit card customers. The data consists of about 100,000 individual customers with data on 10 attributes, including one indicating whether the customer was delinquent. Beginning with the problem definition, we will go through the various steps involved in the data science process as described above.

Step 1: Problem Definition

In this case, the problem is easy to define. Predict the value of the delinquency indicator.

Step 2: Data Cleaning and Feature Engineering

Data cleaning and feature engineering is an important part of the process in our case. The reason behind this is that the data is imbalanced, meaning that it does not have an equal representation of delinquents and non-delinquents.

In fact, the data has 93% non-delinquents, which is expected in the real world as most people do not default on their credit card debt. This imbalance can affect the choice of model and performance metric used. It will also affect the quality of the model.

Step 3: Model Building

Step 4: Performance Measurement

Don’t Miss: How To Find Out Who Owns A Website Domain

Aim: Do You Think A Professional Degree Will Have An Edge Over Numerous Online Courses And Moocs Available For Data Science Enthusiasts

Professor Ramanathan: We see several students approach us with this question. We always illustrate to our students that if their primary motive is to learn, it does not matter what course format is being pursued. In such a scenario, the individual is left with the burden of finding the correct course with sufficient content and information. On the other hand, a professional degree programme is backed by a trusted institution and brings content curated to suit industry requirements and ensures professional partnerships to benefit students.

Giving It All A Meaningful Purpose

Even though some people still expect magic results from AI, it is not magic. It is also our responsibility as members of the AI community to continuously explain what AI is. As Galileo was saying The book of nature was written in the language of mathematics. And mathematics has always helped to model and describe the behavior of the physical world. So to some extent, there is nothing new in AI, its another mathematical tool to describe the nature that surrounds us and the physics law of the world.

On the other hand, there is something special about AI, that makes people ask: Can AI help with the big challenges of our time?

If I use Schneider Electrics domain expertise as example, AI is like a bridge to a greater efficiency. It can boost and maximize our decarbonization and electrification efforts. Accelerating the sustainability gains and finding new solutions to address climate change through data science and analytics makes AI an extraordinary tool at our disposal.

You May Like: How To Register My Own Domain

What Is Data Science

Before we answer why, we have to understand what data science actually is. According to Wikipedia Data science is a multi-disciplinary field that uses scientific methods, processes, algorithms, and systems to extract knowledge and insights from data in various forms, both structured and unstructured, similar to data mining.

Simply put data science is a field where data in its raw form is processed into information.

Domain Knowledge In A Specific Field: Is It Important For A Data Scientist

DOMAIN KNOWLEDGE: Is it Important for a Data Scientist?

Is it important for data scientists to have domain knowledge in a specific field? In this article, Ill answer this question.

As you might have heard from me many times, data science is at the intersection of three big fields:

  • Business thinking
  • Focusing on business thinking, we can break it further down into two big topics:

    • a general business sense: how do you approach a problem, how do you estimate ROI, how do you prioritize, etc.
    • domain knowledge: specific expertise in a given field

    domain knowledge is important

    And domain knowledge is important. You cant come up with meaningful conclusions, and drive results from your data science projects, if you dont know the business you are in. Thats sort of self-evident.

    But how important it is exactly and how much domain knowledge should you have before you apply for a specific data position? In this article Ill explain everything.

    Note: this article is also available in Youtube video format here and in podcast format here

    Don’t Miss: Can I Use My Own Domain With Wix

    Aim: What Was The Impact Of The Covid Pandemic On The Data Science Education Market How Can Governments Nudge More Students To Data Science

    Professor Ramanathan: The pandemic and the work-from-home setting have provided the bandwidth and opportunity for professionals to pursue online courses. Of course, with several data science courses and opportunities available in the market, it may almost appear like the professional is being compelled to take the course. But I feel the situation has simply allowed more time and clarity to choose to upskill.

    In terms of government policies like NEP are instilling significant realisation that we need to encourage multidisciplinary environments. However, to address modern-day problems, we need a well-thought combination of STEM and other disciplines. It will contextualise problems and solutions and help our students stand out.

    Importance Of Domain Knowledge In Data Science

    Domain knowledge is that area of data science which is hardly discussed compared to other areas like programming skills, visualization skills, algorithms or statistics.

    In data science, having domain knowledge will ofcourse differentiates you from other experts of the field, but it wont entirely differentiates you even, because having domain knowledge is different in terms of delivering work, handling out resources and procedure.

    Domain Knowledge

    Domain knowledge is knowledge of a specific, specialized discipline or field, in contrast to general knowledge, or domain-independent knowledge.

    The role of a domain expert is much more specific, the domain in here will be data science so they must be having proficient knowledge in every processes that happens inside this domain. From the technique in selecting data to the technique of processing of that data, a domain expert should be well experienced in every step of process.

    Responsibilities of a Domain Expert

    Importance of Domain Knowledge in Data science

    It is difficult for anyone in random to come up with project ideas in a domain without enough experience and knowledge, it will further be difficult to determine what type of data will be useful for that project. One will need to have experienced vision over the structure and purpose of a project, should have knowledge in what types of variables might be related to the expected outcome so that to be sure in gathering right type of data.

    How to gain Domain knowledge?

    Recommended Reading: Does Google Host Email Domains

    Best Domains For Data Science

    When you learn data science, you should always learn it so that you can use your data science skills in any field. But sometimes you find that you can put all of your interest into one area which is known as domain expertise. When you are in the best position to use your data science skills in a particular area, that area will become your area of expertise. To become a data scientist, you first need to focus on your data science skills so that you can work in any area, but to get the domain expertise, you need to work on the practical use cases. So what are the best domains for data science? In this article, I will introduce you to the best domains for data science based on the average salary you receive in India.

    Is Domain Knowledge Important For Machine Learning

    Domain Knowledge Data Science

    If you incorporate domain knowledge into your architecture and your model, it can make it a lot easier to explain the results, both to yourself and to an outside viewer. Every bit of domain knowledge can serve as a stepping stone through the black box of a machine learning model.

    Nate Rosidi

    Developing machine learning models involves a lot of steps. Whether youre working with labeled or unlabeled data, you might think numbers are just numbers, and it doesnt matter what each of the features of a dataset signifies when it comes to spitting out insights with the potential for true impact. Its true that there are tons of great machine learning libraries out there like scikit-learn which make it straightforward to gather up some data and plop them into a cookie-cutter model. Pretty quickly, you might start to think theres no problem you cant solve with machine learning.

    Frankly, thats a beginners mindset. You are not yet aware of everything you dont know. Datasets given in machine learning courses or the free ones you find online have often already been groomed and are convenient to use when applying machine learning models, but once you take your skills and knowledge out of the play-pen and into the real world, youll face some additional challenges.

    If you start working in areas like outlier detection, which isnt such an everyday human task, the importance of domain knowledge quickly becomes apparent.

    Don’t Miss: How To Make Money With Domains

    Where Is Domain Knowledge Useful

    You may have understood from the above example that domain knowledge is best useful in feature engineering. Feature engineering is creating features using the domain knowledge to optimize the machine learning algorithms.

    Lets see an example in the economics related data to support what we have seen so far. The combination of economics and mathematical concepts is called Econometrics and machine learning particularly regression is being used widely these days to create insights using the raw data. We see two models, one without feature engineering and one with feature engineering using domain knowledge of economics. To keep this blog simple and concise we will only fit the models and compare them.

    For example, lets take the Catalonia GDP data which you can download here. Lets see the head of the dataset

    We will then apply linear regression before and after applying domain knowledge.

    Without application of domain knowledge

    From Domain Expert To Data Scientist

    Heres a common case:You are already an expert in a given domain, you want to stay in this domain and become a data scientist in it. But you dont have any data skills yet.

    I have good news: thats the best and easiest position to be in.Why? Because the majority of the things youll have to learn, in this case, are hard skills.

    Heres a concrete example.

    Lets say that you are a chemist. You have a Chemistry BSc degree, and five years experience at a pharmaceutical company where you have worked as a research assistant. You havent worked with data yet but you have a strong domain knowledge.

    Thats awesome because to move towards data science, youll have a pretty straightforward roadmap:

  • Learn statistics!
  • Okay, I wont go through the whole roadmap here, but if you want to learn more about it, check out my free mini-course called How to become a data scientist.

    The point is that yes, learning coding and statistics is challenging, it will take a lot of time and energy from you But its still not as difficult as getting five years of experience and domain expertise in the pharma field. In other words, the most difficult part is already done.

    So if you have domain knowledge in a specific field already, you want to stay in that field, and you can learn data-science-related hard skills, youll stand out compared to others when applying for positions.

    Also Check: How To Buy A Domain Forever

    Working With A Domain Knowledge Expert

    You may be thinking to yourself, But my background is data science! How exactly am I supposed to get to a point where I understand the business context without years of experience behind me?

    Its a valid concern. The important thing is to remember that you are part of a team and that different voices and experiences enrich the process by bringing new ideas and perspectives to the table. You arent supposed to magically absorb everyone elses expertise you just need to find better ways to tap into that expertise.

    Put it this way: you are surrounded by other people with domain knowledge. Everyone in your company that works on the business side has built up that understanding over the course of their careers. They need your skills to make sense of data to help them further their business goals and you need their contextual understanding to enrich your models. The key is to involve them in the process, right from the planning stage.

    In other words, you need to speak to existing domain experts and stakeholders among your colleagues and ask them to help you define the task. You can then figure out which toolset will best help you solve the problem they describe or get the answers they need, and work together to select the most significant and representative data.

    The Ideal Applicant Vs The Most Applicants

    Why Domain Knowledge Is Important For Data Scientist?

    Lets start with the obvious:If you already have the fundamental data science skills and good domain knowledge in a specific field, thats a huge advantage! Also, sort of unique.

    Lets say:

    • youve learned Python, SQL and statistics from online courses
    • you have a good sense of business
    • and you are fascinated with cars

    And you have your lucky day, too: a few entry level data scientist positions open up at BMW, Tesla and Toyota, right in the city where you live. Perfect match! Awesome! You just send in your CV and are practically hired.

    But the thing is that quite often, youll apply for data positions in a different domain than you are good at right now.

    Maybe there are no openings in your preferred domain in the city you live in. Maybe you want to try something new. Maybe you realize that you could work on more exciting things in other fields in the long term. Or maybe you dont have domain experience in any fields yet because this will be your first job.

    These are realistic and pretty common scenarios. So if youve recognized yourself, dont worry, just keep reading.

    Don’t Miss: How Do You Set Up A Domain

    Aim: How Does An Industry Partnership Add Value To The University Courses

    Professor Ramanathan: Industry partnerships are crucial to educational institutions. The two key components of a data science course are the fundamental conceptual foundation laid by highly qualified academicians and industry stalwarts with on-ground expertise and visibility. Both ensure that the key takeaways are beyond theoretical knowledge and include practical insights and understanding.

    More articles

    Popular Articles