Explaining Big data
What if tell you to organize and analyze data of citizens of New York on Facebook, Instagram and Twitter. It would be insane…. why? Because I am talking about data of 19.49 million of people of what they like, share, comment on, messages, watch videos of and their routine activities. This is a lot of activity and means a lot of data.
On average there are 2.5 quintillion bytes of data processed each day which is growing at a positive rate. With 2 billion people active on Facebook (almost more than quarter of world population) is the largest social media platform.
We are talking trillions and billions of data, so it is obvious that a group of specialists needed to analyze this amount of data and therefore it is called Big Data and people who analyze these data are called Big Data Analysts.
For example, if a food chain wants to open its branches in foreign countries too, it will need ample amount of data to be analyzed of people who it wants to target.
They will organize and analyze data according to religion, cultural preference, weather, seasons, festivals and other geographical and cultural factors. All this colossal amount of data can be analyzed by team of data analysts.
1] Challenges
We can describe the challenges of Big data analytics by three Vs that are Volume, Velocity and Variety.
- Volume: The amount or bulk at which data is streaming. For example, the data of each car on road for governmental measures.
- Velocity: The rate at which data is coming. It can change for example credit card operations data of each customer or downloading on mobile of each customer.
- Variety: The diversity of data; data can be structured, semi-structured and non-structured.
2] Types of Data analytics
There are three ways of data Analytics that are descriptive, predictive and prescriptive.
- Descriptive: The descriptive technique is just the summary of the data, the simplified data. The billion of streaming data is explained in the simple means.
- Predictive: This technique is the main and most important one by analyzing the massive streaming data you will predict the outcome based on your observation. Remember this is used to tell the outcome that might happens in the future not will happens.
- Prescriptive: This technique is beyond the prediction, it derives multiple solutions and prescribed the best solution. For example, the best place to place oil company’s petrol pump for maximum profits. The other two components in this system are feedback and actionable data.
3] Tools to analyze
Big Data analytics can be learnt through courses on internet and the main tools that are used in data analytics are Tableau Public, OpenRefine, KNIME, RapidMiner, Google Fusion Tables, NodeXL, Wolfram Alpha, Google Search Operators, Solver and Dataiku DSS. These tools perform visualization, mining, processing and statistical modeling of data to help analysts with their work.
4] Future of Big Data
The current demand is high of Big data analytics and with technological advancement and availability of internet growing the field will diverse and expands.
The car manufacturers are using Big data of traffic for building future self-driving cars. The medical research laboratories are using Big data to match DNAs compatibility for better use of medicines. The field is growing and its future prospects are bright.
Join 25,000+ smart readers—don’t miss out!
