GRACE AI

In General terms how a chatbot works User's input is processed by a software function called a “classifier” , this classification associates an input sentence with an “intent” (a conversational intent) which produces a response.Think of a classifier as a way of categorizing a piece of data (a sentence) into one of several categories (an intent). The input “how are you?” is classified as an intent, which is associated with a response such as “I’m good” or (better) “I am well.” There are 3 ways to create text classification; Pattern matchers Algorithms Neural networks Regardless of which type of classifier is used, the end-result is a response. Like a music box, there can be additional “movements” associated with the machinery. A response can make use of external information (like weather, a sports score, a web lookup, etc.) but this isn’t specific to chatbots, it’s just additional code. A response may reference specific “parts of speech” in the sentence, for example: ...

In our second scrum meeting, we went over the initial configurations such as git configuration, python virtualenv installation and downloading NLTK lib on our local machine, and we also discussed our project blueprints and the work flow chart. (pls see the flow chart on graceai.blogspot.com) Second sprint, we are going to break down our project to small pieces. Therefore we can start creating milestones and relevant issues on our project tracking system on git lab. On the other hand, we need to create a scenario for our chat bot to achieve and collect more accurate data from users input. Also this will also save some time for us to complete primitive version of our bot. Our first scenario will be a flight reservation bot. Our primarily focus is to find right entities on each sentence and be able to train our Ai to prevent any miscommunication. We are not planning to use any 3rd party APIs . This will be our proof of concept. We are also not going to look for queries on each sentenc...

Creating a Chat Bot with with ML/DL techniques Purpose of our research : In our first meeting, we have talked about what is grace and its capabilities such as how it gets queries and analyzes them. We already implement NLTK, Sci-Kit learn and Tensor-flow libraries in Grace (Grace Search Engine app) so we want to take this to next step. Our aim is to improve our sentiment analysis component, therefore machine can start understand what users say and be able response to them. In this research we are planning to accomplish, a very primitive version of a high intelligence level of chat bot. System Structure Current Structure    General Idea for chat bots Our current scenario isn't ready yet. Second sprint meeting we will be discussing what the scenario would be. Scheduling and Team Organization As it has done previously our discussions will continue on Grace Slack group. and Google groups / gracesearchengine. Our repository will be on gitlab.com/grace...