Category: Chat

  • Wie funktionieren Chatbots und Voicebots?

    Was passiert eigentlich im Hintergrund, wenn man einem Chatbot oder Voicebot eine Frage stellt und eine Antwort erhält? Um die technischen Zusammenhänge zu verstehen, musst du zunächst wissen, was Intents, Utterances, Responses und Entitäten sind.

    Die Antworten eines Bots werden in Intents organisiert

    Ein Intent ist eine im System gespeicherte Nutzerabsicht. Sie besteht aus dem Namen des Intents, den Fragevariationen (Utterances) und natürlich der Antwort (Response):

    Utterances and a Response of an Intent

    Wie du in der Abbildung siehst, stehen bei den Utterances mehrere Variationen der Frage „Wie ist das Wetter in Würzburg?“, bei der Response jedoch nur ein einziger Satz. Der Chatbot kann daher als Antwort immer nur denselben Satz auf die Frage ausgeben.

    Das Erstellen mehrerer Utterances ist notwendig, weil jeder Nutzer seine Frage ein wenig anders formulieren kann. Als Antwort reicht in der Regel eine Response aus. Du kannst aber auch hier Variationen erstellen. Ob diese sinnvoll sind, hängt vom entsprechenden Use Case ab. Die Gesamtanzahl aller Intents bei einem Chatbot ist davon ebenso von dem Use Case abhängig.

    Was für Arten von Intents gibt es?

    Bei den Intents lassen sich zwei verschiedene Arten unterscheiden: Die Intents, bei denen die Antwort des Bots immer gleich ist, und die, bei denen die Antwort individuell angepasst werden muss.

    So lautet die Antwort auf die Frage „Wie lange dauert ein Praktikum?“ immer „Ein Praktikum in unserem Unternehmen dauert 6 Monate“. Dabei ist es irrelevant, ob der Nutzer sich in Berlin oder München aufhält und ob er die Frage heute oder erst morgen stellt.

    Auf die Frage „Wie ist das Wetter?“ kann die Antwort aber nicht immer gleich sein. Das Wetter ist ändert sich täglich und ist darüber hinaus ortsabhängig. Der Bot muss daher erfassen können, wo sich der Nutzer befindet, um eine an die aktuelle Wetterlage angepasste Information zu präsentieren.

     

    Entities helfen dem Bot die richtigen Informationen auszuspielen

    Wenn wir uns nun die Utterances genauer anschauen, werden wir feststellen, dass bestimmte Wörter eine Schlüsselfunktion erfüllen. Sie ermöglichen es dem Bot in der Antwort die richtigen Informationen auszuspielen:

    Diese Schlüsselwörter werden als Entities bezeichnet. Mit ihrer Hilfe kann der Chatbot erkennen, auf welche Informationen er zugreifen muss. Die erste Entity „Weather“ signalisiert dem Bot, dass er auf eine Datenbank wie Open Weather zugreifen muss, um dort die Wettervorhersage abzurufen. Die zweite Entity „Würzburg“ sagt ihm, für welchen Ort er die Vorhersage braucht.

    Ablauf des Antwortprozesses bei einem Chatbot oder Voicebot

    Was passiert nun, wenn Nutzer:innen dem Chatbot die Frage „Wie ist das Wetter in Würzburg?“ stellen?

    Der Chatbot gleicht die Anfrage mit sämtlichen Utterances im System ab und berechnet für jeden Intent einen Confidence Score. Dieser besagt wie groß die Wahrscheinlichkeit ist, dass bei dem jeweiligen Intent die Antwort richtig ist. In unserem Fall ist der Confidence Score bei dem Intent „Weather“ am höchsten.

    Dann untersucht das System die Anfrage auf mögliche Entities:

    Aufgrund dieser beiden Entities weiß der Chatbot, dass er auf eine Datenbank wie Open Weather Maps zugreifen und die Vorhersage für die Stadt Würzburg abrufen muss. Diese Informationen werden dann in die Response eingebaut:

    The words

    Diese Antwort wird dann den Nutzer:innen angezeigt.

    Bei einem Voicebot kommen am Anfang und am Ende noch ein Schritt dazu: Die von den Nutzer:innen über gesprochene Sprache eingegebene Anfrage wird über eine Spracherkennungssoftware (STT =Speech-to-Text) in geschriebenen Text umgewandelt, bevor sie analysiert werden kann. Die Antwort des Chatbots wird über eine Text-to-Speech Software in gesprochene Sprache umgewandelt.

    Du möchtest noch mehr über Chatbots und Voicebots wissen? Dann lade dir unsere Infografik Chat- und Voicebots für Anfänger herunter.

    Der Beitrag Wie funktionieren Chatbots und Voicebots? erschien zuerst auf BOTfriends.

  • Conversation is the new UI

    Over the last decade, users of technology have experienced the benefits of an ever-expanding developer toolkit dedicated to enhancing the interfaces that we use to communicate with our devices and applications. Over the last decade, the typical user interface (UI) has moved well beyond the constraints of the keyboard and mouse or trackpad toward an increasing reliance on touch and voice. In the 2020s, I would like to re-frame an evolving idea: the new frontiers of UI will increasingly leverage modes of artificial intelligence—specifically conversations—to power interactions between human and machine.

  • How Chatbots can Change the World

    As a startup, our dream has always been to be able to change the world and make it a better place. As cliché as it sounds, that is really what drives most startups to dredge on, even when the going gets tough, like really tough! And whilst many of us will not make it to the news, small steps do add up.

    So, can chatbots change the world? And if so, how?

  • Designography 101: How to design user-friendly interfaces

    Designing a user-friendly interface that can cater to your entire target audience domain is not as simple as it seems. We often play with different design approaches and analyse user interactions in our websites/products. This user-behaviour testing helps us to improve the interfaces and enhance the UX. And then the entire cycle repeats.

  • How Automating Our Deployments Workflow Made Our Team More Productive

    Deployments used to be a nightmare. Traditionally, the phases of deployments could take several hours or even days. But in the era of CI/CD, companies are shipping out product updates in minutes. You’ve probably heard of the term CI/CD being thrown around quite a lot lately. In simple terms, it’s a combined practise of Continuous Integration and either Continuous Delivery or Continuous Deployment.

    As a growing startup, we started with no CI/CD at all and after some time, we realised that it caused a lot of headaches for everyone involved. Every time there was a new change, we had to manually build new images, test, update deployment files and manually deploy. It also did not help that bugs kept on coming, and hours were spent in maintaining deployments and keeping production running.

    Needless to say, we have improved a lot since then. After roughly 600 automated deployments, here’s what we learned about CI/CD:

  • How We Used Contentful to Build a Chatbot That Helps With Everyday Legal Problems

    Have you been laid off? Are you behind on your bills? Need to renegotiate a contract?

    In any of these situations, you may need legal information. And if you live in British Columbia, Canada, you can chat with Beagle to get it. (Try it—click the Messenger chat icon on the bottom-right of this page.)

    Tangowork is the technical partner and chatbot consultant to People’s Law School on the Beagle project, with Agentic Digital Media providing design expertise. People’s Law School is a non-profit society in British Columbia dedicated to making the law accessible to everyone.

    For our chatbot technical stack, we chose Contentful to manage the content and Rasa to power the chatbot AI. In this post, we discuss the reasons we chose Contentful for content management and the way we approached integration.

    Contentful: One system to manage content across channels

    A good chatbot requires a tremendous amount of content. Beagle’s team of lawyers spent several months writing and editing over 3000 replies across 71 topics.

    We realized that the new chatbot content was so approachable and easy to read that it would be a shame if we didn’t use it on the website as well. But no one liked the idea of managing content in two places. So we started to explore the concept of “headless” content management systems — systems that manage just the content, leaving you to decide where that content is used.

    We examined Ghost, headless Drupal, headless WordPress, and a handful of others. Nothing seemed to be as mature and purpose-built as Contentful. We ran some experiments with it, liked what we saw, and decided to move forward.

    Using Contentful for the chatbot

    We use Rasa, an open source chatbot engine, to analyze input and decide how to respond. Rasa doesn’t generate language — all possible responses have been written in advance. Rasa simply returns the name of a template to use. Then we retrieve that template from Contentful.

    For example, in response to “i got laid off”, Rasa returns laidOff. Then a piece of middleware, built by our development team, calls the Contentful API and requests the content associated with laidOff. Contentful responds with the content, and then we format the text and send it to the chatbot:

    Content is delivered from Contentful straight to the chatbot

    Using Contentful for the website

    The chatbot response to “i got laid off” works perfectly as the introductory paragraph to a webpage. The chatbot buttons, such as “What to know”, show as website headings. Individual chatbot speech bubbles show as website subheadings and paragraphs. (Read more about the website project in How you can grow your website traffic 51%—overnight—by radically increasing speed.)

    Content is identical in both the chatbot and the website — it’s just reformatted

    Certain pieces of text aren’t appropriate for both the chatbot and the website. For example, we use emoji more liberally in chat. For those instances, we surround the text in question with a span tag like this: <span class="chatOnly">Cool, eh? 😎</span>. The text is then suppressed in the channel where it doesn’t apply.

    The Contentful editing interface is easy for content authors to understand

    Changing text in Contentful instantly updates both the chatbot and the website

    Unified content management in Contentful makes high-quality content easier

    People’s Law School has a very high standard for content. After multiple drafts by their legal content developers, content passes to a plain language reviewer, who performs another check to ensure it’s as clear and easy to read as possible. Finally, it’s reviewed for accuracy by one of a host of external lawyers.

    If this legal review process had to happen twice — once for website content, and once for chatbot content — it would be unsustainable. Instead, Contentful allows us to manage a single set of content, and publish it in two places: on a website and on a chatbot.

    And it doesn’t need to stop there. We can theoretically produce PDFs or other printed collateral via Contentful. Indeed, we’re using maximum-resolution images in Contentful’s media library in case we go down that path. (Contentful resizes and downsamples images on the fly for delivery to the website and chatbot.)

    The lawyers at People’s Law School are thrilled with Contentful as their unified content management solution. “Contentful is fast, easy to use and flexible,” says Drew Jackson, lawyer and project lead. “There’s nothing we’ve tried to do with it that it couldn’t handle. And I’m relieved that when the law changes, we don’t have to update both the website and the chatbot. We change the content once, and it updates everywhere.”

    Chris McGrath

    Chris McGrath

    Founder, Consultant

    Chris loves working on elegant software solutions to difficult problems. He often speaks on topics related to websites, intranets and chatbots. He co-founded ThoughtFarmer, an intranet platform used worldwide. He has a terrible poker face, a mediocre golf swing and a penchant for hasty decisions. On the plus side, he pulls a mean espresso, speaks two obscure languages (Serbo-Croatian and St. Lucian Creole), and never says no to an invitation to go kiteboarding.

    Tangowork Chatbot Consulting

    Tangowork’s consultants use research, design and technology to help your organization create better digital experiences. We specialize in intranets, AI assistants and high-performance websites.

    The post How We Used Contentful to Build a Chatbot That Helps With Everyday Legal Problems appeared first on Tangowork: Consultants for Intranets, AI Assistants, Fast Websites.

  • The Combined Power of Chatbots, Predictive Analytics, and Sentiment Analysis

    Over the last several years, the topic of automation has gotten a great deal of attention. It’s fairly widely accepted that we are in the beginning stages of what has been called the Fourth Industrial Revolution. AI’s ability to automate many routine tasks is at the heart of the notion of the Fourth Industrial Revolution.

    In parallel, I’ve been having some interesting conversations with clients and contacts about the use of predictive analytics, in the form of sentiment analysis, in chat channels. One way to think about this is trying to tap into the power of prediction by using AI to pick up on weak signals of sentiment or preferences that customers might indicate within a conversation (in this case via chat or messenger services).

  • Chatbots for Learning: Expectations vs Reality

    As most educators, trainers and facilitators are now having to move their classes online, they are also looking for better ways to engage their virtual learners. eLearning, live Zoom classes, educational videos are digital tools that might already be used. While live online classes will provide the teacher-student interaction that is very much needed, time limitations exist, just as in physical classrooms. eLearning and videos provide learners with the ability to learn anytime, anywhere, on any device, but these tend to be very much one-sided, with the learner passively watching and navigating through the online content.

  • 5 Powerful Ways Chatbots can Elevate the Customer Experience

    Chatbots are totally changing the way that businesses are engaging with their consumers. 

  • Running an Intranet Workshop: Sharing the Agenda (or “Hero’s Journey”)

    Our intranet workshops usually last from one to three days. Regardless of the length, we like to set everyone’s expectations by sharing an agenda. Although we share it in advance via email, discussing it again at the start of a meeting reminds everyone of the plan and provides an opportunity to make any necessary adjustments.

    Lately I’ve been using the “Hero’s Journey Agenda” to set the stage for my Content Migration workshops. (Read about it in detail on Gamestorming). The Hero’s Journey captivates the audience and prepares them for what lies ahead.

    Hero’s Journey for Intranet Content Migrators

    1. Draw a big circle on a whiteboard. Write “Ordinary Life” at the top, and a little figure that represents a person in your group.
    2. Explain that every epic tale involves a hero that gets the call to leave his or her ordinary life. Write “The call” around 2 o’clock on your circle.
    3. Explain that as the hero embarks on his or her journey, they meet friends to help them along the way. Draw a little figure of “helpers” — perhaps with capes or wizards’ hats — and explain who is going to help your group on this journey.
    4. Draw a big line through your circle and divide it into “Known” and “Unknown”. Explain that we’re crossing into the Unknown, and we’re going to be facing Problems and Perils. Write “Problems & Perils” at 4 or 5 o’clock, and draw some dark clouds and monsters.
    5. At 6 o’clock, draw The Pit. Explain that this is the low point, when everyone is going to be frustrated. And that this might be a good time to break for lunch.
    6. In epic tales, the Hero emerges from the Pit with hope: he or she has acquired “Special Powers”, which you’ll write at 7 or 8 o’clock with lightning bolts, light bulbs and exclamation marks. Explain that the group is going to acquire special abilities, such as the ability to quickly add, edit and manipulate content, files and photos.
    7. Finally, write “The Return” around 10 o’clock. This is where the hero crosses back into the Known world, bringing back Gifts to his or her Ordinary Life. Draw little boxed gifts, and explain that everyone will be returning home as Content Heros. They’ll bring back the gift of Content Migration Know-How, which will benefit others on their team.

    I used this agenda with a credit union a few weeks ago. The intranet manager loved it when I drew her with a cape, my colleague loved it when I drew her with a magic wand, and the content migrators loved thinking of themselves as heros. When we reviewed our takeaways at the end of the day, many people commented that the Hero’s Journey was a highlight!

    The great thing about the Hero’s Journey agenda is that it prepares your group to face a period of frustration during the workshop. This is usually the part where they are learning something new and difficult, or are unravelling a complex problem. But the agenda also assures them that they will break through, learn something new, and leave the workshop with new skills. Presenting the agenda in this way can infuse excitement into what otherwise could appear to be a mundane topic, like content migration.

    Chris McGrath

    Chris McGrath

    Founder, Consultant

    Chris McGrath has led intranet projects for over 20 years. He co-founded ThoughtFarmer, an intranet platform used by hundreds of organizations worldwide, and has worked on many SharePoint projects as well. Now, through Tangowork, Chris provides vendor-neutral intranet strategy and development services.

    Tangowork Chatbot Consulting

    Tangowork’s consultants use research, design and technology to help your organization create better digital experiences. We specialize in intranets, AI assistants and high-performance websites.

    The post Running an Intranet Workshop: Sharing the Agenda (or “Hero’s Journey”) appeared first on Tangowork: Consultants for Intranets, AI Assistants, Fast Websites.