Otys, being a major recruiting software company, encounters a wide variety of nationalities. This comes with certain obstacles, such as different customers speaking different languages which not everyone understands. Thinking about how to overcome this obstacle, Otys thought of a solution featuring Google’s Cloud Translation API.
Otys has the option to send support tickets, which can be done by any customer at any time. This is where the Google Translate functionality comes into play. Once a support ticket has been sent, that support ticket now has the option to be translated automatically. This functionality can be used by clicking the Google Translate button at the top right corner of the support ticket. Clicking this button detects the language of the ticket, retrieves your current profile language and translates the ticket to your current profile language.
This whole process is realized with the use of Elasticsearch, a RESTful search and analytics engine to store and retrieve data in an efficient and expandable way, and Google’s Cloud Translation API, a language detection and translation API which supports over 100 different languages. When the button is pressed, a call to the Elasticsearch storage is made, checking if the translation has already been made. If so, that translation is taken and presented to the user. If not, the translate API is called to translate the text, the translation is stored in the Elastic Search storage and is then presented to the user.
If users reply to the ticket after having the Google Translate feature enabled, they are able to do so in their own preferred language as the reply is automatically translated to the original ticket’s language once it is sent.
Currently, the language detection and translation section of this thesis are performed by using Google’s Cloud Translation API. However, is this the best option for Otys? To answer this question, the research topic focuses on the alternatives of Google’s Cloud Translation API. The research will cover the speed, accuracy and price of every alternative with a critical comparison in the end. Using this gathered information will reveal which alternative suits Otys best.
If you want to cite this thesis in your own thesis, paper, or report, use this format (APA):
Lauwers, J. (2019). Responding to support tickets in an unfamiliar language .
Unpublished thesis, Hogeschool PXL, PXL-Digital.