INTRODUCTION: by providing labeled datasets and in unsupervised

INTRODUCTION:

Artificial Intelligence is method of making the computers
to think intelligently, in the same way humans does. We human can learn, think,
act and make decisions for the given problem. in same way artificial
intelligence powered system can learn think and make decisions of their own.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

 

Machine Learning is a branch in Artificial intelligence
which specifies that the machine intelligence is achieved by learning
(training), same way humans learn to improve their knowledge. Machine learning
is applied in many problems as solutions such as text, games, spasm, images,
videos and speech. It uses different types of learning methods. In supervised learning
the algorithms are trained  by providing
labeled datasets  and in unsupervised learning
,the algorithms are not supervised that is the datasets are not labeled later
the learned data are used to take task-specific decisions.

 

Deep learning is subset of machine learning which is
inspired by the functions and structures of brain and it deals with use of neural
networks. It uses large number of processor to train. The neural networks are
trained using large dataset and a program tells the layers how to behave to the
users. This project is about building a chat bot which provide human like
response and we will discuss about how the deep learning models are used for
training chat bot on social media conservation data.

 

 

OVERVIEW:

 Chat bots are used
in many fields to reduce human works and to provide detail information without
any obstacles. Chat bots are designed to provide instant reply for the
questions that the users asked for.

It can handle hunters of conservation in parallel and are
designed for task specific. The enquiry chat bot play a major role, which
provide information for specific tasks .chat bot can reduce human loads. Each bot
are fed with huge dataset and these datasets will let the bot to decide and
response with quite good answers. Users may feel relaxed while using these chat
bots.

 

This project is all about creating a friendly chat bot.
It provides friendly conservations with user and reply with related answers for
the questions it asked for. This conservation bot can reply the users with
related answers. The conservation is based on datasets that the system is
trained and there are no limitations in defining the dataset. Here the dataset
should be in 2 way conservation i.e. from message and reply message e.g. (hi,
how are you ? ) is the request (I am fine) is the response.

 

CHALLENGES:

Users do not expect complex functionalities, they need
results in short. It is about satisfying their needs. The most challenging
thing is to build a better user interface which is more convenient to the
users. User satisfaction is the most important factor in software fields. There
are many challenges while building a chat bot.

User way of texting:

Many people will have their own way of typing style, it
could be a short sentence or long sentence or more elaborate sentence. Users
can follow any of this style that they want. So building a chat bot to understand
the users typing style is more challenging thing.

User tone of language:

Some people will be unique, their way of talking, usage
of slang, frequent use of certain words, nature of misspelling certain words,
usage of short and so on. So the chat bot should deal with this.

Recognizing user needs: the bot should understand what the
users are trying to say and reply with proper response. Identifying the users
need and processing the replies is most challenging factor.  

 

PROBLEM STATEMENT:

 

 

 

 

 

 

 

OBJECTIVE:

The main objective is to build a chat
bot that perform human like conservations through text. This bot provide a
friendly conservation between machine and user. It understand the users input,
search for the related answer and response it. These responses are achieved
with help of deep learning which involves sequence to sequence model (seq2seq).
This model consist of two main components an encoder and a decoder. The encoder
job is to encapsulate the user input into intermediate representation
and the decoder job is to decode the representation into source data.

 

SCOPE:

1. To build a friendly
chat bot that responds the users.

2. To give meaning full conservation.

 

 

 

 

 

 

 

 

 

 

 

 

  

 

 

x

Hi!
I'm Harold!

Would you like to get a custom essay? How about receiving a customized one?

Check it out