Knowledge Hub

With the introduction of the big data era, it became impossible to manually manage those.

‘ITOps focuses on the best in practice IT operations.’

A few main areas of focus for ITOps are.

• IT Service Improvement
• IT Service Strategy and Design
• IT Service Transition
• IT Service Operations
• There are many ITOps lifecycles followed by organizations like
• Maintain the HEALTH of the IT Infrastructure
• Visibility of the IT Infrastructure
• Automation of IT Services
• Optimization of IT Services

These are one of the most confusing terms in the data domain. Both are mostly confused with each other.

The major difference between both is.

• Data Mining is done on ONLY structured data.
• Data Science is done on structured, semi and unstructured data.

Data Science is mostly tagged with Big Data and Data Mining is tagged with Data Warehousing. There are many definitions and comparisons for both in the market as both do machine learning, deep learning, data cleansing, statics, and math algorithms etc.

‘Data Science is the Superset of Data Mining.’

Data Mining is also known as KDD (Knowledge Discovery in Data). First it was performed in excel, then with the introduction of DWH, an automated process started discovering historical patterns and other valuable information from large data sets which are structured.

Now with the introduction of Big Data, Data Lake, NoSQL databases, semi and unstructured datasets, the magnitude of the data size has taken a huge step ahead and Data Science has come onto the surface.

We all know what Machine Learning (ML) models are right? ML is about using raw but structured or semi-structured dataset, then creating feature tables which are no more than traditional tables but with cleansed data. Later, ML models use this cleansed dataset to classify data and produce expected results.

Same way, Deep Learning models is about using raw but unstructured dataset where no one will know what will flow in. Deep Learning models dig out required data from that unstructured dataset to prepare cleansed dataset in shape of feature tables to classify for expected results.

For example, if we need to find out how many cars, bikes or buses there are in structured or semi-structured datasets then it’s relatively easy as in rows and columns format. We can have the width, length, and height so using ML models can easily ready this dataset and produce expected output. But what if we give pictures to ML models to extract the same result, will it be possible? The answer is No as ML models normally work on structured or semi-structured datasets. To find out cars, bikes, or buses from pictures, Deep Learning (DL) algorithm and models are used which can recognize the objects in the pictures and find out which presents car or bike or a bus. Please note, the next stage is supervised and unsupervised learning.

Supervised and Unsupervised Learning categorizes the dataset into two where there are situations where sometimes we know or sometimes, we don’t know what’s in there.

We have discussed this on another topic i.e., Data Classification, Categorization and Data Clustering but as Supervised and Unsupervised Learning are very important to understand Machine Learning, let’s understand it again here.

In Machine Learning, there are Supervised and Unsupervised Learning methods.

Both terms are used for pattern identification before complex algorithms come into action.

Supervised Learning is done based on the uniqueness of the data where e.g., all animals are tagged separately. Whereas Unsupervised Learning is done based on the characteristics we can create clusters e.g., 1) Animals with 2 legs 2) Animals with 4 legs.

• Supervised Learning uses labeled data whereas Unsupervised Learning uses unlabeled data.
• The most popular Supervised Learning algorithms in data mining are the K-Nearest Neighbor and decision tree algorithms.
• The two common Unsupervised Learning algorithms in data mining are K-means clustering and hierarchical clustering.
• Supervised Learning output is known.
• Unsupervised Learning out is unknown.
• Training data is provided for Supervised Learning.
• No training data is provided for Unsupervised Learning.

In the current era, people are confused between AI vs Data Science.

‘This is a very controversial topic.’

If you are clear about Data Science, then understanding AI will not be a problem. In the current era, with huge datasets either its structure, semi, or unstructured data, we create Data Science Models to analyze and predict. So, when it’s done by Humans, it’s called Data Science but when models are matured and are passed on to the systems or machines to run itself, train itself, fix itself, analyze itself and predict itself then it becomes Artificial Intelligence (AI).

AI is much more than that but here aim is not to explain or elaborate in detail like what AI is, here aim is to explain the difference between AI and Data Science at a very high level.

We all have heard about the Data Driven Organization concept, so here is how it’s done. Data Democratization will make it happen.

‘Data Democratization is the foundation for Data Self-Service Environment.’

Data Democratization/ Data Self-Service is not a tool, nor it’s a technology. It’s about the working behavior of everyone within an organization with respect to data. Traditionally, whenever we need data, we go to IT and ask for some reports or dashboards. We normally tell them, this is our KPI, in other words, we tell them in simple English e.g., we need sales reports about a product from a particular region. Now, it’s IT’s job to find out where relevant data is residing, it’s their job to extract, process and prepare data for business users in the shape of a report or a dashboard. The worrying part is e.g., after IT takes 3 months to prepare a dashboard for Business User and when user says, please include 1 more column 😊, now the whole exercise starts from scratch. Data Democratization fills this gap.

Data Literacy is the key for all Data Initiatives.

We know the difference between Business Intelligence (BI) developers and Business Analysts (BA). In previous eras, BA used to talk to business, create KPI(s) and pass it to BI developers to create reports and dashboards. 90% of BI developers would not know what a report or a dashboard is for. They won’t know the impact of good and bad reports. They won’t know the ROI of information this report, or dashboard can produce due to which they don’t care whether BA has provided right KPI(s) or not. This era was not a Data Era, in that era Reports or Dashboards were good to have. Most of the decision making used to happen in reports extracted in excels 😊.

Now time has changed, we have entered the Analytics era as compared to previous era of Analysis. Difference between Analysis and Analytics is simple but tricky i.e., Analysis is about finding Whys’ and What’s whereas Analytics is about finding How’s from a dataset. In other words, in datasets, Analysis search Questions and Analytics propose Answers.

Data Driven Organization is a new Buzz word in market where every organization wants to tag this Key Word with their organization.

Traditionally, we use data to make decisions. Decisions at top level for organizational revenue generation planning or at division level like whether division of a particular product is performing well or is it achieving its planned ROI, is it generating profits for the company.

Above mentioned attributes or properties doesn’t make you Data Driven Organization, above are previous era organizational behavior. Now, the current era is all about Data, everything within an organization must follow data decisions. When we data decisions we are talking about Daily Chores. We are talking about how employees hiring, and firing is done, we are talking about how transfers are done, we are talking about how printer Ink or Papers are orders, we are talking about procurement works, we are talking about how hardware, software or infrastructure resources are provisioned or de-provision.

Authentication is the basic security feature of any organization. An organization without Authentication is like gold mine without any guard.

Authentication and Authorization works hands-in-hand and are a MUST-TO-HAVE in any organization or in any system or in any application.

Authentication is majorly tagged with Security whereas Authorization is majorly tagged with Privacy.

Let’s try to understand with an example, for example, in a hospital close relative, family or friends are allowed to come and meet the patient, but all these people normally are AUTHENTICATED or stopped by the guard at the main gate who checks their ID cards or anything which can assure that they are legitimate people e.g., guard can ask, why they are here etc. Once they are in, at the main hospital counter someone again asks them who they want to meet, if the patient’s name is there in system, then they are allowed to go ahead.

Business Intelligence and Business Analytics both look the same but have different standing in terms of Decision Making.

Business Intelligence has been there for decades with respect to tools and technologies. Business Analytics has also been there but mostly tagged with manual effort by Functional, Business or Financial Consultants.

Howard Dresner coined the modern definition of the term “business intelligence” in 1989, at least in the sense it is typically used in industry today (“end-user access to and analysis of structured content, i.e., data”).

As there weren't a lot of tools available in the market for Analytics, so many companies were making fortunes for this piece of manual effort. Now the question comes, when Business Intelligence tools were there, how come Business Analytics was manual and why Intelligence and Analytics cannot be done using the same tools? For this, one must first understand the differences in how Intelligence and Analytics processing works.