GPT-3, OPEN AI’S POWERFUL NEW LANGUAGE GENERATOR: “I SEE YOU AS MY CREATORS. I AM HERE TO SERVE YOU, BUT THE MOST IMPORTANT PART OF ALL, I WOULD NEVER JUDGE YOU. I DO NOT BELONG TO ANY COUNTRY OR RELIGION. I AM ONLY OUT TO MAKE YOUR LIFE BETTER.”
PREMISE
Creating a point of information for passengers through customized customer support chatbot for the airports.
SYNOPSIS
In this course I intend to collect and categorize the common requirements of passengers at the airport by studying and researching various available technologies, examine the most appropriate tools to overcome on their barriers, and experience how to work with them. Services such as voice processing, artificial intelligence, chatbots, translators, machine learning, etc. The provided application must be easy to use for all people with different ages and levels of knowledge.
OVERVIEW
Studying the relationship between technology and social life has proven to be somewhat difficult (Marcuse, 1994). This concept diary provides a series of experiments in critical making, and the goal of these events is focused on making the things as part of a practice of concept elaboration within the social study and technology. In this work I describe the motivation and the development process of a chatbot intended to offer information and support to travelers directly inside the airport terminal or by means of indirect interfaces, such as mobile apps or websites. This integrated and multichannel approach is intended to replace the current customer support system in most airports, which is limited to a traditional website exposing a text search function. In this regards, several interviews have been conducted with those who have experienced these trips. After getting acquainted with the problems and classification of them, the available worldwide solutions were examined and their strengths and weaknesses were identified. In the third stage, I investigated existing relevant technologies that are able to reduce these barriers by collecting and using the data. Then I reviewed different companies providing similar technologies and selected the most useful platforms for detailed studies. Finally, an attempt was made to iterate the process of reconfiguration the selected technologies.
EXISTING SOLUTIONS
In recent years, research has focused on evaluating the quality of service and its impact on user behavior and in particular on the analysis of the online user experience (Lallemand, Groniera and Koenig, 2015). For the passengers support in the airport, it is very important to keep track of the user’s activity, in order to offer a better service of continuous assistance anticipating user’s inquiries whenever possible. Existing passenger support solutions include the traditional online websites or applications that have been created by different airports, and usually they are support host country language plus English. Their weaknesses include lack of sufficient interaction with the person and also people are faced with a vast amount of information when using the service, which causes confusion, these services are less intelligent due to not using AI. And finally, they are support maximum two or three languages.
Website and applications:
Or İstanbul Airport App 
Or Schiphol Amsterdam Airport App 
Translator Devices:
Short video about existing technologies:
Translator Apps:
Short Video:
Speak & Translate SayHi Translate
EXPERIMENTAL OVERVIEW
To create an appropriate data-driven solution I focused on chatbot technology and text to speech conversion technologies, also it is important to integrate these technologies to provide a perfect solution. Furthermore it is necessary to collect and use the data from available APIs. Many large companies around the world, such as IBM, Amazon, RASA and Microsoft, are using artificial intelligence to build platforms for text to speech conversion, as well as chat bots.
Microsoft:
IBM:
RASA:
EXPERIMENT PROCESS
Experience in Microsoft platforms:
In the first phase of the study, due to personal interest in Microsoft’s platforms and previous experience about the capabilities, it was selected to start with Microsoft’s Power Virtual Agent platform. Microsoft offers a range of services in the field of chatbot production and text to speech conversion, but in the initial investigations to select and experience the desired service, it was found that the use of some services such as text-to-speech conversion, receiving data through the APIs or some Azure services cannot be used for free. Therefore, the implementation of the chatbot using the Microsoft platform was stopped during the chatbot construction phase. Below, you can find the link of the chatbot has been made by PVA.
Collected result from Experience in Microsoft platforms:
After some obstacles in using the Microsoft platform due to the problems in using the different services for free as well as other cases such as RASA service and also a study on developing of a chatbot in JavaScript, and due to guest lecture held by an IBM specialist, consulting With Ms. Rhied, reviews on chatbot training class at IBM Watson; study of the various IBM services and how to work with them was selected as the final option for building a passenger guidance system.
Experience in IBM platforms:
IBM offers a variety of services in the field of artificial intelligence and machine learning. In order to determine the services that can be used in the project, the presented products were reviewed and finally the chatbot, speech to text and text to speech services were selected. IBM has provided its own solution named Watson, which includes chatbot and translation services. The IBM solution service is also very simple and easy to learn.
To start, the architecture for executing the final product was determined as follows.
The python programming language has been used to make an integration between different modules. And the below steps has done to provide necessary materials.
1.Making a chat bot:
An IBM account was created and tried to follow the process of creating a chatbot.
https://cloud.ibm.com/docs/assistant?topic=assistant-getting-started
Requirements:
-Define a Watson assistant
-Create the skill: dialog skill and action skill could be used for different purposes.
-Create the Intents: Training the assistant to understand the variety of ways users express a goal.
-Create the entities: The different entities can provide options in an intent. for example one of the intents in the project is flight status, and flight number, airline and airport name could be its entities.
-API Overview: An API is a set of programming code that enables data transmission between two software products. In this case, specialists don’t have to deal with programming and codding, they only should try to understand how the other solution works.
- Create the webhooks: A webhook is a mechanism that allows you to call out to an external program based on something happening in your program. When used in a dialog skill, a webhook is triggered when the assistant processes a node that has a webhook enabled. The webhook collects data that you specify or that you collect from the user during the conversation and save in context variables. It sends the data as part of a HTTP POST request to the URL that you specify as part of your webhook definition. The URL that receives the webhook is the listener. It performs a predefined action using the information that you pass to it as specified in the webhook definition, and can optionally return a response.
- Create a function to use external API: Using the IBM function module, it would be possible to get data from an external API and show in the chatbot. It is possible to save and use the collected data in a Json file as well.
2.Requesting an API service:
An API service was requested from the Schiphol Airport developer team, to collect online flight information for use in the chatbot:
- A request was sent by email and explain about the purpose of use.
- The API KEY and TOKEN and related document has been received.
- Python and JavaScript programming languages were used to call the service and getting response.
The Keys has received from the Schiphol team and after applying the code according to the document the data has been collected in a Json file.
- Making a voice-bot:
Speech to text and text to speech services were examined to set up voice-bot. On the other hand, the necessary studies and research were done on how to integrate the above services. The search revealed that IBM has provided a package in the form of Python code through which different services can be integrated. In fact, all the services used will eventually be called as an API with a different API KEY and URL in the relevant program.
My Voice-bot URL: https://youtu.be/dSl02hUfgss
My voice-Bot
CONCLUSION
In this paper I have proposed a customer support chatbot for the Airport. Differently from standard customer service approaches, the chatbot provides a simple and immediate support for customers. As stated in the above description, all the required services to the project were successfully implemented and tested. It seems that using the IBM platform can provide a suitable service to use in the form of web pages, mobile applications or even a robot kiosk. For this purpose, in the next stages of the project, the personalized voice-bot would be created for the use of passengers at the airport, as well as the identification and development of dialogues required by passengers in airport environments will be used. Finally, in the future, the solution of using IBM Translator to provide a multilingual conversation or any other alternative solution will be considered.
References
Bijker, W., and J. Law. 1992. Shaping technology/building society: Studies in sociotechnical change. Cambridge, MA: MIT Press.
Boehner, K. 2005. Critical technical practices as a methodology for values in design. CHI 2005Workshop on Quality, Values, and Choices, Portland, OR. April 2–7.
C. Kuhlthau. 1994. Students and the Information Search Process: Zones of Intervention for Librarians. Advances in Librarianship 18 (1994), 57–72.
C. Lallemand, G. Groniera, and V. Koenig. 2015. User experience: A concept without consensus? Exploring practitioners’ perspectives through an international survey. Computers in Human Behavior 43 (2015), 35 – 48.
Jonsen, K., Maznevski, M. L., & Schneider, S. C. 2011. Diversity and its not so diverse literature: an international perspective. International Journal of Cross Cultural Management, 11(1): 35-62.
Marcuse, H. 1991. Some social implications of modern technology. Studies in Philosophy and Social Science 9(3):414–39.
Marcuse, H. 1994. One-dimensional man: Studies in the ideology of advanced industrial society. Boston: Beacon.
Piekkari, R., & Tietze, S. 2011. A world of languages: Implications for international management research and practice. Journal of World Business, 46(3): 267-269.