BOTfriends Preview

Welcome Intent / Welcome Message

--> to the BOTwiki - The Chatbot Wiki

The Welcome Message is the very first message a chatbot sends when the user is confronted with the chatbot. It serves to pick up the users of a chatbot, provide information and explain what the next steps together can look like.

The first message has the motto: "Create a shared understanding!".

Two different types are distinguished: Either the chatbot sends a message proactively to the potential user or the user initiates communication with the chatbot himself by writing to it.

Relevance of the Welcome Message

First impressions count. This is also true with chatbots. That's why it's so important to put a lot of thought into how you build that message. You should give the user a good feeling right from the start when interacting with the user. The welcome message can be compared to the menu of a website. It offers the user navigation and orientation.

Recommended structure of the start message

The welcome message should contain the following components:

  • Presentation of the chatbot (also an explanation that it is a chatbot)
  • Avatar, Logo (Does my chatbot have a character and a look?)
  • Explain the capabilities of the chatbot (What information can the chatbot provide and where are the limits of the application?)
  • Explanation of how to interact with the chatbot (free text or guided conversation)
  • Explanation of whether and how a Human Handover works

Repeated Playback of the Welcome Message

In some cases it may be useful to replay the welcome message. For example, when new features are implemented, or to remind users of the chatbot's capabilities. Of course, the welcome message can also be played back time-controlled, so that users receive the message again after a longer period of inactivity to receive information about new features or changes.

> Back to BOTwiki - The Chatbot Wiki


BOTfriends Preview

Rasa

--> to the BOTwiki - The Chatbot Wiki

Rasa is a conversational AI platform that provides open-source software, the Rasa Stack, for creating contextual AI assistants and chatbots. In addition to the free Rasa Stack, there is also the Rasa Platform. This is also based on the Rasa Stack, but offers additional features for enterprise customers, such as a user interface with functionalities like the Training Data, Admin, Conversations and Models API. Rasa is developed by the Berlin-based start-up of the same name, and there are also over 300 contributors who promote Rasa in the open-source sense. [ 1][2]

Structure of the Rasa Stack

The Rasa Stack is divided into Rasa NLU and Rasa Core. These are structured in such a way that they can be used completely independently of each other. Thus, it is possible to build only part of the architecture on Rasa and to include additional services. Despite this fact, the two components are very well coordinated and are therefore quickly configured. The Rasa NLU takes over the tasks of intent recognition and entity extraction, while Rasa Core handles the complete session management, context handling and bot responses.[3]

Advantages and disadvantages of Rasa

First of all, the advantage of the open-source idea must be mentioned. On the one hand, one is not dependent on licences that are subject to a fee, and on the other hand, a very high degree of use-case-specific configuration is possible by viewing the entire source code. For example, industry-specific terms such as "cash" and "balance", which have the same meaning in the banking environment but must be differentiated in general, can be mapped much better in a chatbot. Of course, this also means that you have to work more intensively to get an optimal solution when you compare it to ready-made solutions such as Dialogflow from Google or LUIS from Microsoft.
In addition to ready-configured solutions, these also offer hosting of the service. Unlike these cloud solutions, Rasa runs on premise. This gives you complete control over the data generated by the chatbot, but you also have to provide a powerful server. Another advantage is the possibility to use Rasa completely offline. This means that it can be used in offline systems, for example IoT devices.[4]

> Back to BOTwiki - The Chatbot Wiki

Sources

[1] https://www.gartner.com/en/documents/3879492
[2] https://rasa.com/about
[3] https://rasa.com/docs/
[4] https://rasa.com/docs/


BOTfriends Preview

Wizard of Oz Experiment

--> to the BOTwiki - The Chatbot Wiki

In the field of human-computer interaction, the Wizard of Oz method refers to a research experiment in which subjects interact with a computer system that is considered autonomous by the subjects but is actually operated or partially operated by an unseen human.[1]

Chatbots and Wizard of Oz

Chatbots are enormously suitable for the Wizard of Oz experiment. In this way, a use case can already be examined for its"chatbot suitability" before implementation. The findings can then be used to iteratively expand existing flows and define new communication strands. In addition, the collected data, such as utterances, can be used directly for the Chatbot Training be used. A mature human handover tool is even able to automatically convert the tested data into a chatbot.

Recommended procedure

  1. Define different chatbot flows
  2. Integration of a live chat or an empty chatbot that only triggers a human handover.
  3. Manual response to user queries
  4. Derive chatbot flows
  5. Answering the queries based on the chatbot flows
  6. Iterative revision and expansion of the chatbot flows
  7. Manual or automated transfer of the flows into a chatbot builder

> Back to BOTwiki - The Chatbot Wiki

Sources

[1] Kelley, J. F., "An empirical methodology for writing user-friendly natural language computer applications". Proceedings of ACM SIG-CHI '83 Human Factors in Computing systems (Boston, 12-15 December 1983), New York, ACM, pp. 193-196


BOTfriends Preview

RCS - Rich Communication Services

--> to the BOTwiki - The Chatbot Wiki

Rich Communication Services (RCS) is a protocol of mobile operators and is promoted by the GSMA, the industry association of international mobile operators.[1]

RCS and Chatbots

RCS Business Messaging is the evolution of mobile messaging that increases and improves communication between people and businesses. It offers businesses the opportunity to increase customer engagement by using business messaging through chatbots and artificial intelligence (AI). No longer is it necessary to download multiple apps, users are given direct access to a range of brands and services within the messaging app itself, allowing them to work with virtual assistants to book flights, buy clothes, make restaurant reservations and more.[2]

Rich Messaging Chatbot via "SMS"

Simplified, RCS can be seen as the successor to SMS and MMS. Only with rich messaging content such as buttons and videos. RCS is integrated by Google and Apple into the already pre-installed Android Messages[3] app and iMessage [4] app. Thus, every Android and iOS user theoretically has access to RCS.

RBM and ABC

From Android, the system is called RBM (Rich Business Messaging)[5]. Apple calls its service ABC (Apple Business Chat)[6]. To use the systems, however, the transmission protocol RCS must be supported by the respective carrier.

> Back to BOTwiki - The Chatbot Wiki

Sources

[1] https://www.gsma.com/futurenetworks/rcs/
[2] https://www.gsma.com/futurenetworks/rcs/rcs-business-messaging/
[3] https://messages.google.com/web/authentication
[4] https://support.apple.com/explore/messages
[5] https://jibe.google.com/business-messaging/
[6] https://www.apple.com/ios/business-chat/