In today's world, every reader or social media user has different choices/hobbies in terms of reading. For example, if any social media user is searching for a book to read without any specific idea of what s/he wants, s/he wastes a lot of time browsing around on the internet and crawling/trawling through various sites hoping that s/he might get good book. To avoid confusion, the authors are building a recommendation system for every reader/user that helps to recommend a book based on his choices, hobbies, or what s/he had read previously that will be massive help for users instead wasting time on various sites. Data from social media is the powerful fuel that can be used to helps in decision making and building a recommendation engine. Social media data in the different format is biggest challenge for the business to ingest data at the reasonable speed and further process. In social media data, it is difficult to detect and capture data. Real-time recommendation engine for users, which includes data ingestion methods, challenges, metadata problem, analysis, and consumption, is discussed here.
TopBackground
In today’s world, every reader or social media user has different choices/hobbies in terms of reading. For example, if any social media user is searching for a book to read without any specific idea of what s/he want, s/he waste a lot of time browsing around on the internet and crawling/trawling through various sites hoping that s/he might get good book. To avoid confusion, we are building a recommendation system for every reader user that helps to recommends book based on his choices, hobbies or what s/he had read previously that will be massive help for users instead wasting time on various sites. Data from social media is the powerful fuel that can be used to helps in decision making and building a Recommendation engine, Data of social media in the different format, it's the biggest challenge for the business to ingest data at the reasonable speed and further process, in social media data it is difficult to Detection and capture of changed data. Once data is available, it is difficult to prepare data for analysis, transform (cleansing and normalizing the data) and convert data into structured format. In this Chapter, we are describing a case study about Real Time Recommendation Engine for users which includes data ingestion methods, challenges, metadata problem, analysis and consumption.
What is Recommendation Engine
Recommendation Engine is a black box which analysis some set of users and recommends the items which a user may like. The analysis uses the user details like Gender, Age, Geographical Location, Online searches, book information, previous hobbies/books readers are interested (Techtarget, n.d.)
Before we learn deeper aspects of recommendation engines. Let’s first understand about the real life and online recommendation engines.
Recommendation engine is a field that involves the, collection, analysis, and presentation of social media data and efficient decision making and recommendation of books of readers The goals of a recommendation engine include:
- ●
Properly recommendation of books to readers based on their interest and offers
- ●
Collecting the data from social media in a sound manner for analytics
- ●
Producing analysis that accurately describe user requirements, interest
Benefits of Recommendation Engine
Following benefits are provided by implementing Recommendation Engine
Figure 1. Benefits of a Recommendation Engine
- 1.
Platform for Recommendation engine using Big Analytics: Once Data (Volatile and static) has been captured, stored and properly preserved in a secure manner, analysis of it can begin. Domain expert/Data Scientist/Business analyst use data mining and analytics tools to extract information from a large set of Library unstructured data from social media, which is then transformed into an understandable structured data for further use Recommendation Engine is providing capability to provide quick recommendation based on search and analyse a mountain of data quickly and efficiently. They can search keywords related to library science such as book name, journals, interest etc. in a big data clusters in different languages which is beneficial
- 2.
Driving Standardization: In one sense, creating recommendation engine standards in big data is merely a temporary first step on a much longer journey then Analytics for building recommendation engine will give a huge impetus for the standardization of recommendation process in big data domain and library data management. infrastructure. That key shift will inevitably catalyse IT modernization and improve internal IT services maturity, quick way of resolution recommendation issues, effective decision making etc. Also helps in Application security portfolio rationalization
- 3.
Geographic Transparency and Traceability:Big Data (Library’s Data such as book name) and reader's data (such as name, location, is frequently sensitive and its location is therefore important, since data entering or exiting national borders can contravene national and international regulations, such as the EU Data Protection Directive .To address this, big data defines process which is transparent about the geographic location where data and services are stored, and allows clients to keep data on their own servers, using a VPN connection