Welcome to the world of intelligent chatbots, your companion and conversation agents which would make your life smarter. A leading research paper even said by 2020, the average person will have more conversations with bots than with their spouse. So be ready to embrace this new life in a year from now.
Ok hold on, have you ever tried telling Siri or Google to “Find restaurants which doesn’t serve pizza”. At least they are both consistent in some way, they gave the same answer – suggesting restaurants which serve pizza.
Ok how about Sofia, the first citizen humanoid robot, which is making its way to every media event and giving interviews and boost of human like conversations. Well the truth is far from reality, it is providing an illusion of understanding conversation, but as you start asking intelligent questions you would realize it can answer fixed set of questions.
Well by now, you should be able to clear out the noise from reality. So, should I invest in chatbots with all these limitations? Yes, any technology would have its limitations, but you need to be aware of what you can build now, what to avoid and how to work around the limitations. I have seen many companies trying to build sophisticated chatbots using products from leading chatbot vendors and cloud offerings, spending million on dollars and hitting a roadblock.
If you go by what is being projected and start building it out, you would soon realize these limitations one way or the other. The problem is that most of the vendors claim it’s very easy to build a chatbot, but in reality, all of these techniques fall short when it comes to building a true conversational agent.
With current implementations of chatbot, we are probably at the first generation of AI chatbots which are more or less scripted and giving answers to pointed questions. What I mean by scripted is that it is trained to understand general vocabulary, entities, the metaphor, synonyms etc. The chatbot uses fixed set of flows to understand the context. For domain specific use cases, additional training is required, and you need to train on specific domain terminology and relationship between the words. While, there are research going on using deep neural nets, we are still quite far away from building a true conversational chatbot which understands the nitty-gritty of language and domain.
For instance, if you are building a shopping advisor chatbot, the term “black and white dress” implies “black and white” as color and dress as category. You might expect the color “black and white” is fairly generic and should be easily identified by the AI system, but that’s not really the case.
Based on my experience on building a sophisticated shopping personalized advisor, none of the AI NLP implementation fitted the requirements. A simple scenario is these 3 sets of sentences – “black and white dress”, “and black dress” and “blue jeans and white shirt”. In all these 3 examples, the use of word “and” means different meaning. In the first case, its represents a combined color “black and white”, in second instance “and” represent a brand and in third instance two queries joined by a conjunction (i.e. and). Even with required training, a generalizing model was not possible with any of the available solutions. These are just one of the many examples I am highlighting.Imagine the complexity when dealing with medical literature.
In order to get a realistic view of what an AI chatbots can achieve in today’s environment, current limitations and workaround and what to expect in the future, you can refer to my book REAL AI for more details.
Our aim in this book is to provide a real view of AI, where it exists today and what we can achieve in future.
Our aim in this book is to provide a real view of AI, where it exists today and what we can achieve in future. Through this book, we plan to answer these following questions in a vendor neutral way –
For an architecture stack perspective, you have the low powered embedded device installed inside the ball or embedded as part of the design and manufacturing process, its provides at least 6 Axis combo sensor for accelerometer and gyroscopes reading to identify any movement in 3d space. A Motion SDK is installed on top of the device to identify any movements in general and communicate the reading to the cloud. In cloud, we have the learning model or the training data. Basically, we would ask an expert batsman to bat and play various expert strokes like cover drive etc. and record their movements from sensors (bats/pads etc) as well as visuals (postures etc), this would be used as the training / test data and comparison would be made against it. As we are comparing 3D models, machine learning approaches like dimension reduction can be employed ( and many new innovation approaches) to compare two motion and predict the similarity. Similar training data is captured from an expert baller, along with other conceptual information like hand movements, pitch angles etc.
In order to build out the solution, you need to employ computer vision algorithms on the edge. You can build this using commercial available API’s or using various open source deep learning framework like Theano, TensorFlow, Café etc. Deep learning is a branch of machine learning for learning multiple levels of representation through neural networks. Neural network in image processing, automatically infer rules for recognizing images instead of you extracting thousands of features to detect images.
