Data Analytics for Beginners

hello and welcome in this video we attempt to provide you a jog and free introduction to data analytics it's quite a buzz these days and this video is meant for people who are yet to take their first step towards data analytics so let's get started what you see written here on the slide are some of the common terms that are being used very frequently these days especially if you are a part of the corporate world and if you are looking for say different job portals and all you would know that these are the terms which are commonly used and more frequently used these days business analytics business intelligence data analytics and data science it's not important for us to get into a debate as to how these vary but for a starter there is a good overlap between all of these so there is a starting point and when you develop your career in a particular direction you can become more specific about what you want to pursue but just for a quick introduction sake let me tell you business analytics is the use of analytics focus to business that's that's for decision making primarily for managers or stakeholders in the business data analytics is an overarching term which you can say supersedes the scenario of business and goes into the field of healthcare but could could go into a customer research could go into education so data analytics is an overall term business analytics can be a subset of it business intelligence would be a mix of analytics and a lot of analysis which might come from your understanding of the processes and the way your processes flow so this is more to do with analytics plus your subject matter expertise and data science is is I would say the vastest term among all thing is and it's very popular these days this is a mix of multiple skills and we'll talk about that as we move one later parallel to these terms there is another term which is quite popular and as is being used very frequently by people it is big data so what is a big data so let me put it in Templeton's any data that's beyond your systems capacity to process could be a big data for your system when we talk about big data there are certain weeds which are important v as in velocity so it is getting generated at a very high speed V as in volume so there is a large volume of data that's getting stored there is variety of data so it's getting generated to multiple sources the data may not be in a given specific file format or structure all the time it could be from n number of resources and there is something to do with the veracity or you can refer to the accuracy of the data that's been captured so big data is more technical I would say and technical owing to the aspects associated with its storage and the processing speed so the techniques the underlying techniques to work on the big data and approach from the analytic side of it would be the same but yes there are certain technical elements get associated what's the burden what is it that people are talking about so for example Gartner says that in 2018 data and analytics can't be ignored analytics will drive major innovation and disrupt established business models in the coming years technical professionals need to adapt their data and analytics architecture from end to end to meet the demand of analytics everywhere and McKinsey says there will be a shortage of talent necessary for organizations to take advantage of big data by 2018 the United States alone could face a shortage of 140 thousand two hundred and ninety thousand people with deep analytic skills as well as 1.5 million managers and analysts with the know how to use the analysis of big data to make effective decisions now these are some reported names in the field of consulting and research and and if they feel that there is going to be a change in the way the businesses have been operating so far of course the businesses will need to hire for the right skill set and that we're a shortage of the right manpower in the industry is anticipated in fact if you talk about Howard Business Review it is clearly called out that sexiest job of the 21st century is data analytics so you can imagine that this is going to be there this is going to be the demand and the future there for those of us who have not yet taken their first step in this direction should definitely consider doing that now now let's try to understand why the entire scenario has changed so much one is that yes we know there is a demand that has come up but what has led to that demand of course you all know that there is a good increase or explosion in the digital footprint there is more and more social media that has come up every other person has access to a smartphone the digital devices tablets and computers and all of this and in the arena of social media is somewhere or the other getting recorded and and some people refer to it as a digital trace so our access to Internet the Internet of Things and social media all of these put together have led to a complete explosion in the availability of data and to put it in precise numbers this is how the stats looked like so 0.79 zettabytes of data was generated til about 2009 between 2009 to 2011 you can say there is one point two zero bytes of data that's been generated and it continues to increase in fact it's anticipated that by the year 2020 who will have 35 zettabytes of data and those of you were not familiar with zettabytes just assume that it's a lot and lot of data but you can look at almost an explosion here several hundred times compared to what it was there earlier and and no wonder why the availability of data that's being generated would definitely encourage the businesses and their enterprises to understand and extract the value of the data but where we might be lacking the fact that our education system didn't get modified in most of the countries the education system did not change as for the enormous amount of data that's going to be available so people continue to study what they were traditionally studying for years and at the same point in time the businesses where the management was slow to react to the huge availability of data they did not do sufficient investments in their employees to equip them with the right skills and prepare them for the upcoming demand having said that some of you watching this video might be students now there's nothing that you've lost you can still go ahead and build a skill set in the field of data science and there is a long way to go there would be people watching this video who are already working and and maybe they have decent years of work experience irrespective of that you can definitely find a place for yourself in the data analytics arena and we'll talk about it in the following slides so no matter where you are in your career if you're a working professional point is that you can always take that first step and gain something out of it provided you are willing to take that first step now if you say that you're in the middle of your career and you want to survive for next 20 years without acquiring these skills the times could be really challenging for you is all that I can say right now so far we've only talked about how the business scenario has changed and where the challenges could be anticipated if the businesses do not adapt to the understanding of data but what is it that's in store for people who are working are the individuals so IBM predicts the demand for data scientists will soar 28 percent by 2020 and that's a huge demand by all standards now let's have a look kind of salaries that different jobs and the field of data analytics are offered so of course there is a category of data scientists and advanced analytics the one that you see it kind of framed in green here what is the data scientist I just like to touch upon that so they a scientist is basically in amalgamation of three skills number one a coder a person who can write codes knows programming languages and has a comfort with the second would be the statisticians are the people with skills in the area of mathematics and third would be the business acumen so data scientist is not just one skill it's an amalgamation of three skills the coders the mathematicians and the business people so you see that a data scientist is a balance between these three skills now it is quite possible that some of us do not have a background into coding can we still make a career into data analytics yes very well if you are really high on your business acumen and you know what you want in mathematical terms you can still go ahead and ask your coders to work for you and they will take care of it so it is not necessary that you need to balance all three you can start from where you are just leverage on your strengths that you have moving on as for a report published by Christ water Coopers companies will look mostly for business people with analytics skills and not just analysts what does it mean well again it points out towards the fact that no matter how good you are with your data but there are certain practical aspects related to a business that you need to understand in order to be able to leverage the information that your data has given you so again re-emphasize in wherever you are in your career if you've made certain progresses in the direction of business you can very well adapt the edge through analytics and then take your business goals forward you can see there multiple categories of jobs so there are data-driven decision makers of course they are highly paid compared to the data analysts and there are analytics managers if you see the last row here these people will always have an edge compared to a plain vanila data analyst let's look at the future so once you enter the data analytics field how does the salary progress you can see the enormous amount of increase as the experience in the field grows and by all standards if you study how the salaries are distributed across different industries and sectors by all standards this is a very comfortable paycheck that we are talking about here and then this is pretty much applicable to the global scenario so irrespective of which country you are in sooner or later this is going to catch up so let's understand what is analytics and like I showed you I will try to keep everything jog and free so first of all the word analytics came into existence from Greece they had a word called analytic O's which meant involving analysis what is the difference between analysis and analytics it's very simple analytics always has an element of data associated with it and analysis has more to do with the procedures and darren's to the procedures in very simple terms definitions of analytics so there are two definitions that I like and we've mentioned the names of the persons who have quoted these definitions the analysis of data typically large sets of data by the use of mathematics statistics and computer software it's important to note here that most of the work that you'll be doing in the field of analytics would be supported by some of the other computer software a programming language or a tool it is almost impossible to work on those large sets of data and and do all the calculations manually the other definition of data analytics which is more application oriented is that analytics is the science of using data to build models that lead to better decisions that in turn add value to individuals companies and institutions so this is a more practical definition and if you just pay attention towards the words mentioned in bold this kind of sums up everything that you need to know and what an analyst does let's get into different types of analytics so there are three types of analytics first one is descriptive this describes what has already occurred it helps the business understand how things are going so this is a path data and you're just preparing a summary of that data and trying to understand this is more reactive in nature and it has mostly to do with the past predictive which is the most popular when we talk about of all types of analytics tells us what will probably happen in the future as a result of something that's already happened so now you're leveraging on the information that's available to you from the past data and trying to make decisions for the future they help the business forecast future behavior and results it is proactive so bases what has happened we want to take certain proactive measures for what's going to happen in future and the last one which is closely relative to predictive is prescriptive analytics it helps the business prescribe right course of action they not only tell us what probably will happen but also what should be done if it happens and it is again a proactive measures so predictive tells us what's going to happen prescriptive is our action that we should take when we know something's going to materialize so these are three simple forms of analytic talkin about the languages and the tools initially the tools to perform advanced analytics were only available while licensing and and the companies used to bear a hefty cost for these the axis was always limited to implies of some large size corporations but now they're ample open source solutions supported by large communities across the globe that are constantly working on these open source solutions and these come quite handy for all data enthusiasts two most popular languages are R and Python both offer integrated Suites of software facilities for data manipulation calculation and graphical display now you'd see there's a lot of debate available on these when you go online which one is better whether we should use R or we should use Python the the simple response without wasting any time on that debate is that both are good choose one and move on we've now been looking at some of the use cases of analytics and and let me tell you even if you're not directly involved in the business of doing analytics yourself you can be assured that your buying behavior your data your presence on social media and through a lot of other channels is definitely being analyzed by the people who you do your business with nobody at this point in time in the digital age is untouched when it comes to analytic our use cases will only emphasize that further 80% of the data that's available today is not in a structured format that as the data does not naturally come into texts and columns it could be a free expression written on internet like a Facebook post text analytics is used to analyze customers sentiments let's say we're in the business of providing technical support to the customers and we provide the technical support through a chat based medium where the customers type their concerns and we immediately on a real-time basis try to cater to their requirements now if say a customer is repeatedly using these words repair not working slow and there could be another set of words trouble frustrating waste refund return escalation now if you see that most of these words have a negative connotation now and there could be exceptions here and there but but overall a customer who's initiating a chat and talking about repairs of course he's facing some or the other problem and so is the case with similar words now looking at this you can very well imagine the sentiment of a customer so in this case you can very well imagine what would be the sentiment of the customer on the other end he would be a customer who is almost given up on you if say for example he's a new customer and he has to face all of this you can very well imagine that even if he survives with your product he is not going to talk positively about you that's guaranteed right so what can we do to to prevent these things from happening and and can we take charge of the situation immediately and do a course kawrench HR analytics now employee attrition if you know is a big problem for the companies and they spend roughly about 20 percent of the c-d-c the cost to company of a churn resource to find his or her back fill and not only that this is just to get the back fill when a back fill takes up that position he needs a learning curve to be able to come to the same level of performance so it's it's it's about money that you spend and the patience that you have to have to bear with the new employee till the time he catches up data scientists have developed algorithm sparked by key indicators that taken together can dramatically predict and employees intent to stay with or leave the company so let's say there are about 25 to 30 attributes that are critical when it comes to predicting implies attrition and if a company constantly keeps an eye on those it would be able to address the employees issues in a more proactive way rather than waiting for the day when an employee comes and submits his resignation and then you try to do an exit interview and then try to retain that employee that that's almost too late he's already made up his mind to move out another case could be a customer lifetime value and I often ask this too while I was teaching people why do you think the companies provide you a loyalty card and a lot of responses come a mixed set of responses people will say that they want to make me feel good I earn some loyalty points and I invest those points there so they get more business out of me of course it's in a way to pay you respect that you are a valued customer but the real model is to be able to track what you're doing see it's easy in terms of an online transaction to figure out that you logged in and you used a particular user ID and password to make a purchase say on an online portal so they know who's purchasing water and they keep a track of it but in a brick-and-mortar store where people just walk in and pick a number of products that they want to buy I say it's it's a supermarket how would you track what's my lifetime value if you do not have a loyalty card so if I have a particular number associated with the final billing that I do at the counter you know that I've logged in after two months and and these are the items that I purchased this could be a good way to track the outcomes of different promotional campaigns that the supermarket's run from time to time so say for example if I gave a very good offer a huge discount on a consumable that is in regular use of my customers I will be able to see if they have been able to take advantage of this is a very interesting case and then let's understand that these while these are specific examples they can be generalized across industries and scenarios here we are talking about India and we are talking about a problem of Tiger what what happens is there are a lot of groups that are into this business of Tiger poaching and they they assume that there's some superstitions or beliefs that exist the tigers bones can be used for making certain medicines which are a solution to certain incurable diseases now this has been happening for years and and these people who are poachers are pretty smart so they know the jungles really well at times better than the forest officers themselves and that's why there is a change between the two all the time so over the past few years computer codes were written and about twenty five thousand data points have been analyzed since 1972 across six hundred and five districts on vine life poaching crimes using this data and applying appropriate techniques the wild lifesavers could narrow down on certain hot spots where the tiger crime was more likely to happen and this helped them prevent and control Tiger poaching now talking about healthcare analytics and this is a very very important and well-known study called Framingham Heart Study it was performed in the United States so it's the name of a town where the people participated in the study it originated in 1948 when the scientists and participants embarked on an ambitious project to identify risk factors now for heart disease today the study remains a world-class epicenter for cutting-edge heart brain bone and sleep research what it enables you to do is that there when you go to the site you'll be able to see that putting your characteristics you can very well figure out what is your risk score to acquire a cardiovascular disease say in a timeframe of ten years so you can generate a Framingham risk score that tells you that you are prone to have a heart disease given the kind of habits that you have right now and it's an interesting area so you can see that it's not just in terms of business applications that we talk about analytics it is not just about targeting the right customer for a promotion it goes on to prevent crimes it goes on to save your life ensure that you live a longer and healthier life it can be applied to almost all the industries and sectors that's why it's fair to say that the world around us is changing and we need to equip ourselves with the right set of skills it's a short video we tried to put together to ensure that you get a sneak peek into what is analytics if you like this video please don't forget to subscribe to our channel that encourages us to continue making more such videos and do recommend us to your friends who might find it valuable thank you for your time

Be First to Comment

Leave a Reply

Your email address will not be published. Required fields are marked *