Descriptive analytics in social media
Descriptive analytics is the modern form of data analytics which is the conventional form of Business intelligence and data analysis. Descriptive analytics seeks to provide a summary of the data in the form of graphs and reports. Descriptive analytics commonly used two main tools, Data Aggregation and Data Mining. It presents past data in simple graphs and numbers that are easily understood by the audience and help a wide business audience. Read more about What does descriptive analytics for social media do?
Social media analytics is the process of collecting and analysing data from social media like facebook, twitter, Instagram, Pinterest. Such data are analysed by data analytics experts Business analytics to track the performance and speed of their product marketing.?Social Media analytics is defined as the art and science of extracting the hidden useful information from the vast amount of unstructured social media data to make precise and accurate decisions.
Data identification and Data Analysis is the two main steps involved in Descriptive analytics. The main questions answered by Descriptive analysis are 5W’s- “Who?“, “What?“, “When?“, “When?“, and “Why?“. these questions help to determine the proper data sources to analyse and answer the essential questions.
Data Identification
Data identification is the process of identifying and analysing the unstructured raw data to focus more on analysis. Raw data is decrypted and interpreted. once these raw data is analysed, we can get lots of information. on a high level, unprocessed data takes a few forms to extract the messages. These include noisy data, relevant and irrelevant data, filtered data, data that display vague messages, data that display the precise information, data that convey the exact message and the real reasons behind them. The derive the exact information from an unprocessed raw data, we need to start processing it, redefine the dataset by including data that we want to extract and organise data to identify information.
In the context of social media, data identification means What and Which content is of our interest. In addition to what and which, we want to know who wrote the content. Where it is found on social media and how many times does the content appeared on that website. Are we interested in particular information or location? When did someone say something on social media?
Further, Descriptive data needs to follow few contexts for extracting useful information from social media.
Structure– Structure data is a data that has been arranged in the form of the database so that its elements can be made more addressable for analysis and processing. The unstructured data are least formatted data and is of no use.
Language– Language becomes a priority if we want to know the sentiments and emotions of the post rather than the number of mention.
Region– it is important to make sure that the data included in the analysis is only from the part of the world where the analysis is focussed on. For example- If we want to know the popularity of our Prime Minister Narendra Modi for the next Lok Sabha elections, then we must make sure that the data collected is from India only.
Venue– Social Media content is getting generated in accordance with the venue such as news sites and social networking sites like Facebook, Twitter, Instagram, snapchat. Depending upon the type and content of data, the venue becomes very important for data identification.
Type of content – The content of data could be in the form of text (text written on social media is easy to read if you know the language), images(drawings, photos, simple sketches), videos(recording, live streams) and audios( audio recordings).
Time– It is important to collect data posted in the time frame.
Data Analysis
Data Analysis is the set of activities that involves transforming raw data into insights, which in turn leads to a new pillar of knowledge and business model. in other words, data analysis is the phase that takes raw data as input and transforms that information into useful content. different types of data analysis that can be performed with social media are demography, analysis of post, sentiments, geography etc. The data analysis process begins once we have evaluate the problem and we have sufficient data to drive meaningful results. If we know that the data is sufficient to draw the conclusion, we need to build a data model.
Developing a data model is the process where we organise data element and compute how the individual data elements are related to each other. In the analysis of our data, it is quite complex to have several tools available at our end. The process involved in data analysis are as follow:
Depth of analysis- Descriptive analytics are based on streaming data, ad hoc analysis in raw data or deep analysis on accumulated data. This analysis is really driven by the time available to come up with the results of the data. This analytics can answer the following questions-
- How many people mentioned their mail account in their posts?
- Which car has the maximum number of sales during Diwali?
- Which company is getting maximum mentions in the context of social business?
The domain of Analysis– The domain of the analysis is mainly classified into two main categories: External social media and Internal Social media. Most of the time when people mention social media, they mean external social media. External social media includes the content from popular sites such as Facebook, Twitter, Instagram, LinkedIn. Internal social media is a private social media to pass communication within the business.
Machine Capacity– Capacity number need to address not only CPU needs but also the network capacity needed to collect and analyse data. Real-time analysis in social media is an important tool when we are trying to understand the public perception of certain topic as it allows for reaction and an immediate change in course. In real time analysis, we assume that the data is processed at the rate that is leas than real-time.
Velocity of Data- The velocity of data in social media is classified into two main category: data at rest and data at motion. Dimension of velocity of data can help us to answer many questions such as: How the sentiments of the voters is changing about the general elections during the course of time? Are the voters conveying positive sentiments about the party leader who is actually winning the election? In such cases analysis is directly correlated to the complexity and performance of the analytical tools or system. A highly complex tool produces more amounts of details. The second type of analysis is the context of velocity of data at rest. This analysis is performed once the data is fully and successfully collected. Performing this analysis can provide review such as: Which product of your company is more successful than that of your competitor? What are the success rate of your business in the early phase after setting your business??