Last time we took a look into what GiftCase was doing through several scenarios covering certain topics. Now, let’s dive into the core of GiftCase, explaining how it works from an architectural point of view.
As we can see depicted above, the most important, core part of GiftCase is the Application Back-end. It is here that the generating recommendation takes place, utilizing algorithms which derive their informed input from several external services. But first, in order for the Back-end to collect informed input, this input has to be properly exposed, meaning it has to come to life through preparation in respective external services.
One such service is the Telco Analysis & Reasoning, which fetches user data from the Telco Network via CDRs (Charging Data Records). Each CDR contains various aggregate, summarized data regarding multiple types of events (such as calls and network sessions) in the Telco Network. Based upon this data from the CDRs the Reasoning Engine containing various data analysis algorithms within the Telco Analysis & Reasoning service provides output – in the form of subscriber info. Consider the following simplified example – extracting two types of information from the CDRs – the online time (i.e. amount of time the subscriber has spent online on the Internet) and traffic amount (i.e. amount of megabytes used). Reasoning upon this information it is possible to classify a subscriber as a:
- Multimedia subscriber – having relatively low online time with a high traffic amount, e.g. watching YouTube videos;
- Non-multimedia subscriber – having high online time with a relatively low traffic amount, e.g. reading a lot of text online.
When queried for such information by the Application Back-end for a specific GiftCase user, the Telco Analysis & Reasoning service provides it through the respective API. Knowing the type of user is important when recommending gifts; a multimedia subscriber is more likely to be satisfied with a movie than a book as a gift.
Similar to how the Telco Network provides CDRs for reasoning and analysis, the social networks (e.g. Facebook) provide a friend list and login features for a specific user. Furthermore, content providing services such as eventful, TasteKid, Google Books and BestBuy provide content and/or information in the form as follows. Eventful provides event information, TasteKid provides a means to fetch similar music/movies/books, Google Books provides book prices and details and BestBuy provides music/movies prices and details. A mashup of all these services is necessary in order for GiftCase to work properly as intended.
Gmail is used as an external service providing push notification functionality. Once user sends a gift, the recipient receives a push notification in the form of an email. Alternatively, the notification is visible within the GiftCase Application Front-end.
GiftCase provides to its users a modern, user-friendly, web-based Application Front-end. State-of-the-art technologies such as HTML5, CSS3, Bootstrap, jQuery and Apache Cordova were put to use during the development phase. Below is a screenshot of the desktop experience, showing the landing (main) page.
Screenshot below depicts the process of viewing recommended gifts, choosing one and successfully sending it to the gift recipient.
Finally, a gift recipient is notified of an incoming gift and is able to accept it or reject it, as depicted below.
Since GiftCase Application Front-end is web-based and is developed with mobile users in mind, it works great with all major mobile operating systems such as Android and iOS.
If you want to know more about S-CASE or try using it today, then you can get it for free right here.