There are a bunch of e-commerce stores taking advantage of chatbots as well. One example that I was playing with was from Fynd that enables you to ask for specific products and they'll display them to you directly within Messenger. What's more, Facebook even allows you to make payments via Messenger bots, opening up a whole world of possibility to e-commerce stores.
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.
When considering potential uses, first assess the impact on resources. There are two options here: replacement or empowerment. Replacement is clearly easier as you don’t need to consider integration with existing processes and you can build from scratch. Empowerment enhances an existing process by making it more flexible, accommodating, accessible and simple for users.
As in the prior method, each class is given with some number of example sentences. Once again each sentence is broken down by word (stemmed) and each word becomes an input for the neural network. The synaptic weights are then calculated by iterating through the training data thousands of times, each time adjusting the weights slightly to greater accuracy. By recalculating back across multiple layers (“back-propagation”) the weights of all synapses are calibrated while the results are compared to the training data output. These weights are like a ‘strength’ measure, in a neuron the synaptic weight is what causes something to be more memorable than not. You remember a thing more because you’ve seen it more times: each time the ‘weight’ increases slightly.
I argued that it is super hard to scale a one-trick TODA into a general assistant that helps the user getting things done across multiple tasks. An intelligence assistant is arguably expected to hold an informal chit-chat with the user. It is this area where we are staring into perhaps the biggest challenge of AI. Observe how Samantha introduces herself to Joaquin Phoenix’s Ted in the clip below:
Once you’ve determined these factors, you can develop the front-end web app or microservice. You might decide to integrate a chatbot into a customer support website where a customer clicks on an icon that immediately triggers a chatbot conversation. You could also integrate a chatbot into another communication channel, whether it’s Slack or Facebook Messenger. Building a “Slackbot,” for example, gives your users another way to get help or find information within a familiar interface.
We need to know the specific intents in the request (we will call them as entities), for eg — the answers to the questions like when?, where?, how many? etc., that correspond to extracting the information from the user request about datetime, location, number respectively. Here datetime, location, number are the entities. Quoting the above weather example, the entities can be ‘datetime’ (user provided information) and location(note — location need not be an explicit input provided by the user and will be determined from the user location as default, if nothing is specified).
ELIZA's key method of operation (copied by chatbot designers ever since) involves the recognition of clue words or phrases in the input, and the output of corresponding pre-prepared or pre-programmed responses that can move the conversation forward in an apparently meaningful way (e.g. by responding to any input that contains the word 'MOTHER' with 'TELL ME MORE ABOUT YOUR FAMILY'). Thus an illusion of understanding is generated, even though the processing involved has been merely superficial. ELIZA showed that such an illusion is surprisingly easy to generate, because human judges are so ready to give the benefit of the doubt when conversational responses are capable of being interpreted as "intelligent".