Furthermore, major banks today are facing increasing pressure to remain competitive as challenger banks and fintech startups crowd the industry. As a result, these banks should consider implementing chatbots wherever human employees are performing basic and time-consuming tasks. This would cut down on salary and benefit costs, improve back-office efficiency, and deliver better customer care.
…utilizing chat, messaging, or other natural language interfaces (i.e. voice) to interact with people, brands, or services and bots that heretofore have had no real place in the bidirectional, asynchronous messaging context. The net result is that you and I will be talking to brands and companies over Facebook Messenger, WhatsApp, Telegram, Slack, and elsewhere before year’s end, and will find it normal.
Do the nature of our services and size of our customer base warrant an investment in a more efficient and automated customer service response? How can we offer a more streamlined experience without (necessarily) increasing costly human resources? Amtrak’s website receives over 375,000 daily visitors, and they wanted a solution that provided users with instant access to online self-service.
I will not go into the details of extracting each feature value here. It can be referred from the documentation of rasa-core link that I provided above. So, assuming we extracted all the required feature values from the sample conversations in the required format, we can then train an AI model like LSTM followed by softmax to predict the next_action. Referring to the above figure, this is what the ‘dialogue management’ component does. Why LSTM is more appropriate? — As mentioned above, we want our model to be context aware and look back into the conversational history to predict the next_action. This is akin to a time-series model (pls see my other LSTM-Time series article) and hence can be best captured in the memory state of the LSTM model. The amount of conversational history we want to look back can be a configurable hyper-parameter to the model.
There is a general worry that the bot can’t understand the intent of the customer. The bots are first trained with the actual data. Most companies that already have a chatbot must be having logs of conversations. Developers use that logs to analyze what customers are trying to ask and what does that mean. With a combination of Machine Learning models and tools built, developers match questions that customer asks and answers with the best suitable answer. For example: If a customer is asking “Where is my payment receipt?” and “I have not received a payment receipt”, mean the same thing. Developers strength is in training the models so that the chatbot is able to connect both of those questions to correct intent and as an output produces the correct answer. If there is no extensive data available, different APIs data can be used to train the chatbot.
Oftentimes, brands have a passive approach to customer interactions. They only communicate with their audience once a consumer has contacted them first. A chatbot automatically sends a welcome notification when a person arrives on your website or social media profile making the user aware of your chatbots presence. This makes you seem more proactive, thus enhancing your brand's reputation and can even increase interactions, having a positive effect on your sales numbers, too.
This means our questions must fit with the programming they have been given. Using our weather bot as an example once more, the question ‘Will it rain tomorrow’ could be answered easily. However if the programming is not there, the question ‘Will I need a brolly tomorrow’ may cause the chatbot to respond with a ‘I am sorry, I didn’t understand the question’ type response.
Die meisten Chatbots greifen auf eine vorgefertigte Datenbank, die sog. Wissensdatenbank mit Antworten und Erkennungsmustern, zurück. Das Programm zerlegt die eingegebene Frage zuerst in Einzelteile und verarbeitet diese nach vorgegebenen Regeln. Dabei können Schreibweisen harmonisiert (Groß- und Kleinschreibung, Umlaute etc.), Satzzeichen interpretiert und Tippfehler ausgeglichen werden (Preprocessing). Im zweiten Schritt erfolgt dann die eigentliche Erkennung der Frage. Diese wird üblicherweise über Erkennungsmuster gelöst, manche Chatbots erlauben darüber hinaus die Verschachtelung verschiedener Mustererkennungen über sogenannte Makros. Wird eine zur Frage passende Antwort erkannt, kann diese noch angepasst werden (beispielsweise können skriptgesteuert berechnete Daten eingefügt werden – „In Ulm sind es heute 37 °C.“). Diesen Vorgang nennt man Postprocessing. Die daraus entstandene Antwort wird dann ausgegeben. Moderne kommerzielle Chatbot-Programme erlauben darüber hinaus den direkten Zugriff auf die gesamte Verarbeitung über eingebaute Skriptsprachen und Programmierschnittstellen.
Previous generations of chatbots were present on company websites, e.g. Ask Jenn from Alaska Airlines which debuted in 2008 or Expedia's virtual customer service agent which launched in 2011. The newer generation of chatbots includes IBM Watson-powered "Rocky", introduced in February 2017 by the New York City-based e-commerce company Rare Carat to provide information to prospective diamond buyers.