How to Build an AI-Powered Virtual Professor Chatbot for Messaging App

May 6, 2024

Introduction

Recently, researchers from Chennai introduced the ‘World’s First Autonomous AI University Professor’, known as the ‘Malar Teacher’. This AI teacher / professor has been built with an understanding of the entire engineering syllabus of Anna University.

The most interesting fact about ‘Malar Teacher’ is that it is accessible through WhatsApp. Since it has access to all the recommended study materials, it can teach any concept from the syllabus, and simplify topics to a 10-year-old's comprehension level.

Leveraging WhatsApp, a widely used messaging application in India, was a masterstroke, as it widely increased the accessibility of this AI, and demonstrated how exactly AI might create a positive impact in India.

In this article, we will demonstrate how to build a similar AI Professor Chatbot, using purely open-source technologies and an open LLM - Mistral-7B. This can be highly relevant for future EdTech startups, universities, coaching institutes and other businesses operating in the Education sector.

Since access to WhatsApp API is through third-party partners, we will instead use Telegram for demonstration purposes. However, the core methodology, especially the AI application architecture used, would be very similar in case of WhatsApp. You would just need to replace the send/receive APIs.

Let’s get started.

Chatbot Workflow

We’ll be using the Retrieval Augmented Generation (RAG) technique for our chatbot. It’s a method of “grounding” an LLM’s response by connecting it to an external knowledge source. This is extremely useful when we want our applications to read and understand documents. By sending contextually relevant information from the documents to the LLM, one can receive information about them by querying the LLM in simple natural language.

Below is a diagrammatic workflow for our chatbot - it will help in understanding the tech under the hood.

‍

Components used:

Mistral-7B LLM hosted on Ollama
Chromadb Vector Store
LangChain for document ingestion and retrieval
Telegram API & Python SDK

Let’s Get to the Code

For hosting the LLM we’ll be needing a GPU-server to handle the AI workload. E2E Networks provides a fleet of advanced GPUs which are tailored specifically for this purpose.

You can check out the GPU offerings by heading over to https://myaccount.e2e networks.com/. Click on Compute on the left-hand side, then click on Nodes, and Add New Node.

‍

‍

First, install all the necessary dependencies.


langchain
python-telegram-bot
ollama
chromadb
asyncio
nest_asyncio

Now open Telegram and search for BotFather to get an API token for the Telegram bot. BotFather is a service offered by Telegram to manage and create new bots.

‍

With the command /newbot you can name and create your own bot. You’ll receive an API_TOKEN of that bot.

Spin up a Jupyter notebook and import all the required libraries:


import logging
from telegram import Update
from telegram.constants import ParseMode
from telegram.ext import Updater, CommandHandler, MessageHandler, filters, ApplicationBuilder, ContextTypes 
from langchain.document_loaders import PyPDFLoader
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.vectorstores import Chroma
from langchain_text_splitters import RecursiveCharacterTextSplitter
from ollama import Client
import nest_asyncio
from telegram.helpers import escape_markdown
import asyncio

Set up the API_TOKEN, the logging info, the Ollama client, and a textsplitter object.

- The Telegram Bot API utilizes asynchronous programming to handle and execute requests concurrently.

- When an `await` statement is encountered, the program jumps to another ready task while the awaited task completes.

- An event loop manages the flow of execution whenever an `await` statement is encountered. It continuously checks for tasks that are ready to run and executes them based on their priority and readiness.

- The event loop takes control and executes other ready tasks when an `await` statement is reached.

- If another `await` statement is encountered, the event loop again takes control and decides the next action.

- Async programming ensures efficient utilization of resources by executing other parts of the code while one part is busy.

- Asynchronous programming is highly effective for Telegram bots handling thousands of users simultaneously.


nest_asyncio.apply()

# Replace 'YOUR_API_TOKEN' with the token you received from the BotFather
API_TOKEN = 'YOUR_API_TOKEN'


# Set up logging
logging.basicConfig(format='%(asctime)s - %(name)s - %(levelname)s - %(message)s', level=logging.INFO)
logger = logging.getLogger(__name__)

client = Client(host='http://localhost:11434')

text_splitter = RecursiveCharacterTextSplitter(
    # Set a really small chunk size, just to show.
    chunk_size=512,
    chunk_overlap=20,
    length_function=len,
    is_separator_regex=False,
)

Refer to this to start a Ollama server and pull the Mistra-7B model on your local server.

Load the embeddings model to convert text into vectors:


embeddings = HuggingFaceEmbeddings(model_name= 'BAAI/bge-large-en')

We create a function to convert PDF into Document format with a chunk size of 512 tokens each as defined by the text splitter.


def load_and_process_documents(file_path):

    loader = PyPDFLoader(file_path)
    pages = loader.load_and_split(text_splitter=text_splitter)
    for page in pages:
        page.page_content = page.page_content.replace('\n',' ')
    return pages

We create a function which will refresh the user data (vector DB) every time the command ‘/start’ is sent. This will allow users to start over fresh.


async def start(update: Update, context: ContextTypes.DEFAULT_TYPE):
    """Handler for the /start command"""
    user_id = update.effective_user.id
   
    context.bot_data[user_id] = {}
    await update.message.reply_text('Welcome! Please send me the PDF documents you want to process.')

Then we create a function to handle the document processing stage. Every time the user sends in a PDF document, it is indexed into the vector DB.


async def document_handler(update: Update, context: ContextTypes.DEFAULT_TYPE):
    """Handler for receiving PDF documents"""
    user_id = update.effective_user.id
    
     # Initialize user-specific data if it doesn't exist
    if user_id not in context.bot_data:
        context.bot_data[user_id] = {}
        
    document = update.message.document
    if document.mime_type == 'application/pdf':
        file_id = document.file_id
        new_file = await context.bot.get_file(file_id)
        file_path = f"{file_id}.pdf"
        await new_file.download_to_drive(file_path)
        
        pages = load_and_process_documents(file_path)
        if 'vectordb' not in context.bot_data[user_id]:
            vectordb = Chroma.from_documents(pages, embeddings)
            context.bot_data[user_id]['vectordb'] = vectordb
        else:
            vectordb = context.bot_data[user_id]['vectordb']
            vectordb.add_documents(pages)
        
        await update.message.reply_text('PDF document received and processed. You can now ask questions about the content.')
    else:
        await update.message.reply_text(f"Unsupported file type: {document.mime_type}. Skipping this file.")

Given a query, this function will create a prompt for the LLM. The prompt will contain the context fetched from the Vector DB, along with the query.


def get_prompt(question, vectordb):
    
    documents = vectordb.similarity_search(question, k=10)
    context = '\n'.join(doc.page_content for doc in documents)
    
    prompt = f"""Using only the context below, answer the following question:
    context : {context}
    question: {question}"""
    
    return prompt

Now we write a function to handle user queries. Note that all the Telegram functions are user specific. This means that for every user there’s a unique vector DB involved (as given by context.bot_data[‘user_id’]). This ensures that the bot can handle concurrent requests from multiple users without mixing up the document information provided by these users. This is because the users’ documents are separately indexed for every user.


async def question_handler(update: Update, context: ContextTypes.DEFAULT_TYPE):
    """Handler for answering questions based on the processed documents"""
    user_id = update.effective_user.id
    question = update.message.text
    vectordb = context.bot_data.get(user_id, {}).get('vectordb')
    if vectordb:
        prompt = get_prompt(question, vectordb)
        
        response = client.chat(model='mistral:instruct', messages=[
          {
            'role': 'user',
            'content': prompt,
          },
        ])

        await context.bot.send_message(chat_id=update.effective_chat.id, text=escape_markdown(response['message']['content']))
    else:
        await update.message.reply_text('No processed documents found. Please send PDF documents first.')

Then we create the main function to deploy the bot.


def main():
    """Main function to run the bot"""
 
    application = ApplicationBuilder().token(API_TOKEN).build()

    # Register command and message handlers
    
    application.add_handler(CommandHandler("start", start))
    application.add_handler(MessageHandler(filters.Document.ALL, document_handler))
    application.add_handler(MessageHandler(filters.TEXT & ~filters.COMMAND, question_handler))

    # Start the bot
    application.run_polling()
   # application.idle()

if __name__ == '__main__':
    main()

Results

I sent this Arxiv paper to the bot. It’s called “Don’t Think about Pink Elephants”, and it’s about the fact that certain AI models, when prompted to not think about certain objects, invariably end up thinking about it - which is a flaw in their design, and is very similar to how human brains work.

‍

Voila! We have created a Virtual Professor chatbot that can read and understand scientific papers and respond to your queries. Of course this chatbot can be used for many other use cases, as the LLM (Mistral 7B) it uses is a general purpose LLM and can understand queries and context in a variety of domains other than just EdTech.

Final Note

You can also implement a similar chatbot in Whatsapp, but that process is a little bit complicated as it requires signing up for a Whatsapp Business account, and then selecting a third-party service provider that offers Whatsapp API integration.

Sign up for Free Trial

Latest Blogs

November 4, 2024

A Complete Guide To Customer Acquisition For Startups

Any business is enlivened by its customers. Therefore, a strategy to constantly bring in new clients is an ongoing requirement. In this regard, having a proper customer acquisition strategy can be of great importance.

So, if you are just starting your business, or planning to expand it, read on to learn more about this concept.

The problem with customer acquisition

As an organization, when working in a diverse and competitive market like India, you need to have a well-defined customer acquisition strategy to attain success. However, this is where most startups struggle. Now, you may have a great product or service, but if you are not in the right place targeting the right demographic, you are not likely to get the results you want.

To resolve this, typically, companies invest, but if that is not channelized properly, it will be futile.

So, the best way out of this dilemma is to have a clear customer acquisition strategy in place.

How can you create the ideal customer acquisition strategy for your business?

Define what your goals are

You need to define your goals so that you can meet the revenue expectations you have for the current fiscal year. You need to find a value for the metrics –

MRR – Monthly recurring revenue, which tells you all the income that can be generated from all your income channels.
CLV – Customer lifetime value tells you how much a customer is willing to spend on your business during your mutual relationship duration.
CAC – Customer acquisition costs, which tells how much your organization needs to spend to acquire customers constantly.
Churn rate – It tells you the rate at which customers stop doing business.

All these metrics tell you how well you will be able to grow your business and revenue.

Identify your ideal customers

You need to understand who your current customers are and who your target customers are. Once you are aware of your customer base, you can focus your energies in that direction and get the maximum sale of your products or services. You can also understand what your customers require through various analytics and markers and address them to leverage your products/services towards them.

Choose your channels for customer acquisition

How will you acquire customers who will eventually tell at what scale and at what rate you need to expand your business? You could market and sell your products on social media channels like Instagram, Facebook and YouTube, or invest in paid marketing like Google Ads. You need to develop a unique strategy for each of these channels.

Communicate with your customers

If you know exactly what your customers have in mind, then you will be able to develop your customer strategy with a clear perspective in mind. You can do it through surveys or customer opinion forms, email contact forms, blog posts and social media posts. After that, you just need to measure the analytics, clearly understand the insights, and improve your strategy accordingly.

Combining these strategies with your long-term business plan will bring results. However, there will be challenges on the way, where you need to adapt as per the requirements to make the most of it. At the same time, introducing new technologies like AI and ML can also solve such issues easily. To learn more about the use of AI and ML and how they are transforming businesses, keep referring to the blog section of E2E Networks.

‍

Reference Links

https://www.helpscout.com/customer-acquisition/

https://www.cloudways.com/blog/customer-acquisition-strategy-for-startups/

https://blog.hubspot.com/service/customer-acquisition

This is a decorative image for: Constructing 3D objects through Deep Learning

October 18, 2022

Image-based 3D Object Reconstruction State-of-the-Art and trends in the Deep Learning Era

3D reconstruction is one of the most complex issues of deep learning systems. There have been multiple types of research in this field, and almost everything has been tried on it — computer vision, computer graphics and machine learning, but to no avail. However, that has resulted in CNN or convolutional neural networks foraying into this field, which has yielded some success.

The Main Objective of the 3D Object Reconstruction

Developing this deep learning technology aims to infer the shape of 3D objects from 2D images. So, to conduct the experiment, you need the following:

Highly calibrated cameras that take a photograph of the image from various angles.
Large training datasets can predict the geometry of the object whose 3D image reconstruction needs to be done. These datasets can be collected from a database of images, or they can be collected and sampled from a video.

By using the apparatus and datasets, you will be able to proceed with the 3D reconstruction from 2D datasets.

State-of-the-art Technology Used by the Datasets for the Reconstruction of 3D Objects

The technology used for this purpose needs to stick to the following parameters:

Input

Training with the help of one or multiple RGB images, where the segmentation of the 3D ground truth needs to be done. It could be one image, multiple images or even a video stream.

The testing will also be done on the same parameters, which will also help to create a uniform, cluttered background, or both.

Output

The volumetric output will be done in both high and low resolution, and the surface output will be generated through parameterisation, template deformation and point cloud. Moreover, the direct and intermediate outputs will be calculated this way.

Network architecture used

The architecture used in training is 3D-VAE-GAN, which has an encoder and a decoder, with TL-Net and conditional GAN. At the same time, the testing architecture is 3D-VAE, which has an encoder and a decoder.

Training used

The degree of supervision used in 2D vs 3D supervision, weak supervision along with loss functions have to be included in this system. The training procedure is adversarial training with joint 2D and 3D embeddings. Also, the network architecture is extremely important for the speed and processing quality of the output images.

Practical applications and use cases

Volumetric representations and surface representations can do the reconstruction. Powerful computer systems need to be used for reconstruction.

Given below are some of the places where 3D Object Reconstruction Deep Learning Systems are used:

3D reconstruction technology can be used in the Police Department for drawing the faces of criminals whose images have been procured from a crime site where their faces are not completely revealed.
It can be used for re-modelling ruins at ancient architectural sites. The rubble or the debris stubs of structures can be used to recreate the entire building structure and get an idea of how it looked in the past.
They can be used in plastic surgery where the organs, face, limbs or any other portion of the body has been damaged and needs to be rebuilt.
It can be used in airport security, where concealed shapes can be used for guessing whether a person is armed or is carrying explosives or not.
It can also help in completing DNA sequences.

So, if you are planning to implement this technology, then you can rent the required infrastructure from E2E Networks and avoid investing in it. And if you plan to learn more about such topics, then keep a tab on the blog section of the website.

‍

Reference Links

https://tongtianta.site/paper/68922

https://github.com/natowi/3D-Reconstruction-with-Deep-Learning-Methods

This is a decorative image for: Comprehensive Guide to Deep Q-Learning for Data Science Enthusiasts

October 18, 2022

A Comprehensive Guide To Deep Q-Learning For Data Science Enthusiasts

For all data science enthusiasts who would love to dig deep, we have composed a write-up about Q-Learning specifically for you all. Deep Q-Learning and Reinforcement learning (RL) are extremely popular these days. These two data science methodologies use Python libraries like TensorFlow 2 and openAI’s Gym environment.

So, read on to know more.

What is Deep Q-Learning?

Deep Q-Learning utilizes the principles of Q-learning, but instead of using the Q-table, it uses the neural network. The algorithm of deep Q-Learning uses the states as input and the optimal Q-value of every action possible as the output. The agent gathers and stores all the previous experiences in the memory of the trained tuple in the following order:

State> Next state> Action> Reward

The neural network training stability increases using a random batch of previous data by using the experience replay. Experience replay also means the previous experiences stocking, and the target network uses it for training and calculation of the Q-network and the predicted Q-Value. This neural network uses openAI Gym, which is provided by taxi-v3 environments.

Now, any understanding of Deep Q-Learning is incomplete without talking about Reinforcement Learning.

What is Reinforcement Learning?

Reinforcement is a subsection of ML. This part of ML is related to the action in which an environmental agent participates in a reward-based system and uses Reinforcement Learning to maximize the rewards. Reinforcement Learning is a different technique from unsupervised learning or supervised learning because it does not require a supervised input/output pair. The number of corrections is also less, so it is a highly efficient technique.

Now, the understanding of reinforcement learning is incomplete without knowing about Markov Decision Process (MDP). MDP is involved with each state that has been presented in the results of the environment, derived from the state previously there. The information which composes both states is gathered and transferred to the decision process. The task of the chosen agent is to maximize the awards. The MDP optimizes the actions and helps construct the optimal policy.

For developing the MDP, you need to follow the Q-Learning Algorithm, which is an extremely important part of data science and machine learning.

What is Q-Learning Algorithm?

The process of Q-Learning is important for understanding the data from scratch. It involves defining the parameters, choosing the actions from the current state and also choosing the actions from the previous state and then developing a Q-table for maximizing the results or output rewards.

The 4 steps that are involved in Q-Learning:

Initializing parameters – The RL (reinforcement learning) model learns the set of actions that the agent requires in the state, environment and time.
Identifying current state – The model stores the prior records for optimal action definition for maximizing the results. For acting in the present state, the state needs to be identified and perform an action combination for it.
Choosing the optimal action set and gaining the relevant experience – A Q-table is generated from the data with a set of specific states and actions, and the weight of this data is calculated for updating the Q-Table to the following step.
Updating Q-table rewards and next state determination – After the relevant experience is gained and agents start getting environmental records. The reward amplitude helps to present the subsequent step.

In case the Q-table size is huge, then the generation of the model is a time-consuming process. This situation requires Deep Q-learning.

Hopefully, this write-up has provided an outline of Deep Q-Learning and its related concepts. If you wish to learn more about such topics, then keep a tab on the blog section of the E2E Networks website.

Reference Links

https://analyticsindiamag.com/comprehensive-guide-to-deep-q-learning-for-data-science-enthusiasts/

https://medium.com/@jereminuerofficial/a-comprehensive-guide-to-deep-q-learning-8aeed632f52f

October 13, 2022

GAUDI: A Neural Architect for Immersive 3D Scene Generation

The evolution of artificial intelligence in the past decade has been staggering, and now the focus is shifting towards AI and ML systems to understand and generate 3D spaces. As a result, there has been extensive research on manipulating 3D generative models. In this regard, Apple’s AI and ML scientists have developed GAUDI, a method specifically for this job.

An introduction to GAUDI

The GAUDI 3D immersive technique founders named it after the famous architect Antoni Gaudi. This AI model takes the help of a camera pose decoder, which enables it to guess the possible camera angles of a scene. Hence, the decoder then makes it possible to predict the 3D canvas from almost every angle.

What does GAUDI do?

GAUDI can perform multiple functions –

The extensions of these generative models have a tremendous effect on ML and computer vision. Pragmatically, such models are highly useful. They are applied in model-based reinforcement learning and planning world models, SLAM is s, or 3D content creation.
Generative modelling for 3D objects has been used for generating scenes using graf, pigan, and gsn, which incorporate a GAN (Generative Adversarial Network). The generator codes radiance fields exclusively. Using the 3D space in the scene along with the camera pose generates the 3D image from that point. This point has a density scalar and RGB value for that specific point in 3D space. This can be done from a 2D camera view. It does this by imposing 3D datasets on those 2D shots. It isolates various objects and scenes and combines them to render a new scene altogether.
GAUDI also removes GANs pathologies like mode collapse and improved GAN.
GAUDI also uses this to train data on a canonical coordinate system. You can compare it by looking at the trajectory of the scenes.

How is GAUDI applied to the content?

The steps of application for GAUDI have been given below:

Each trajectory is created, which consists of a sequence of posed images (These images are from a 3D scene) encoded into a latent representation. This representation which has a radiance field or what we refer to as the 3D scene and the camera path is created in a disentangled way. The results are interpreted as free parameters. The problem is optimized by and formulation of a reconstruction objective.
This simple training process is then scaled to trajectories, thousands of them creating a large number of views. The model samples the radiance fields totally from the previous distribution that the model has learned.
The scenes are thus synthesized by interpolation within the hidden space.
The scaling of 3D scenes generates many scenes that contain thousands of images. During training, there is no issue related to canonical orientation or mode collapse.
A novel de-noising optimization technique is used to find hidden representations that collaborate in modelling the camera poses and the radiance field to create multiple datasets with state-of-the-art performance in generating 3D scenes by building a setup that uses images and text.

To conclude, GAUDI has more capabilities and can also be used for sampling various images and video datasets. Furthermore, this will make a foray into AR (augmented reality) and VR (virtual reality). With GAUDI in hand, the sky is only the limit in the field of media creation. So, if you enjoy reading about the latest development in the field of AI and ML, then keep a tab on the blog section of the E2E Networks website.

‍

Reference Links

https://www.researchgate.net/publication/362323995_GAUDI_A_Neural_Architect_for_Immersive_3D_Scene_Generation

https://www.technology.org/2022/07/31/gaudi-a-neural-architect-for-immersive-3d-scene-generation/

https://www.patentlyapple.com/2022/08/apple-has-unveiled-gaudi-a-neural-architect-for-immersive-3d-scene-generation.html