Smart Bridge Internship Report

Smart Bridge Internship
Project Report on
AI Powered News Search App

By
K Kruthendar Reddy
Malla Reddy Institute of Engineering and

Technology
Hyderabad, Telangana, India
INTRODUCTION
1.1 Overview:
The web is home to massive amounts of data, with more being
created every day. Organizations can harness this constant
stream of information to gain understanding, plan strategies, and
find opportunities. Enriched news data can help your application
make dynamic connections across current events faster.
1.2 Purpose:
The purpose of this project is to develop a user-friendly web
application which makes it easier for the users to find relevant and
recent news articles according to their interests and queries.
All we need to do is give a keyword regarding the required news

to the chatbot so that it can fetch the news articles and links. The
web application also analyses the sentiment present in the article.
LITERATURE SURVEY
2.1 Existing Problem:
The news applications that are currently available can be
confusing, and contain huge data which is inappropriate to the
user to search for a particular article from a substantial amount of
data.
It is also not possible to determine the tone or the sentiment of

the required news using the apps that are available.
2.2 Proposed Solution:

A news search web application which is created by using IBM
discovery service available in IBM cloud.
This IBM discovery service when integrated with Node Red Flow, it
creates a simple, organized, an easy to access user interface i.e.
news hunting application which fetches the user with relevant
news from a vast data since IBM discovery service always updates
itself with the latest news.
By adding sentiment analysis to this application can make the app

UI more unique and user friendly.
THEORITICAL ANALYSIS
3.1 Block Diagram:
The block diagram shows that the user interacts with the UI of the
app created by the Node Red Flow. The UI is integrated with the
Watson Discovery News which keeps updating itself with the latest
news from the internet. This when integrating with slack bot
makes an interactive news bot which fetches the relevant news
according to the query of the user.
3.2 Software Designing:
In order to build this app, the following things are need to be
done:
1. Build a Server Side Application using Node-RED.

2. Use the pre-built Watson Discovery News collection.
3. Access the Watson Discovery Service through the Discovery
API.
4. Use a Slack interface to query the data.
5. Push news alerts out to web notification.
6. Deploy the app on IBM Cloud.
Setting up the development environment
1. Install Git and create a GitHub account. Git installation can be

done through the following website https://git-
scm.com/downloads. Once this is done, open
https://github.com/ and create a git account.
2. Open https://slack.com/intl/en-in/ and create a slack account
and a channel.
3. Open https://cloud.ibm.com/registration and create an IBM
Cloud account. Then start creating a Node-RED application.
4. Create and deploy Watson discovery news app locally.
5. Integrate Watson Discovery Service to the Node-RED
application.
6. Install Node.js from https://nodejs.org/en/download/.
7. Open the slack channel and create a bot application.
8. Integrate slack bot with Watson discovery news app.
The node red UI can be made by using the debug node, function
node, form node, sentiment node and a few chart and label nodes
and connecting them as shown below.
EXPERIMENTAL ANALYSIS
First we create the node red application by creating a flow using 5
simple nodes. This UI is then integrated with Watson Discovery
Service. The user can now enter the query in the UI and get the
relevant news from the Watson discovery (which collects the
data/news from the net and keeps updating itself) and also know
the sentiment of the news (if it’s positive or negative).
Then slack bot is integrated with Watson Discovery Service to
make it more interactive and user friendly where the user can
directly ask the bot to fetch a particular news using the relevant
keywords.
User can also search for the news in the application which will be
available in browser at localhost - http://localhost:3000.
FLOWCHART
RESULT
Slack bot output:
Localhost sample:
Node Red UI output:
This can be seen in the below link:
https://node-red-kiozu.eu-gb.mybluemix.net/ui/#!/0?
socketid=igIl1MQD2cXiFCuiAAAA
Git link:
https://eu-
gb.git.cloud.ibm.com/kruthendarreddy8/NodeREDKIOZU.git
ADVANTAGES AND DISADVANTAGES
 Provides an easy user friendly application to get news related to
the topic that the user wants unlike the available apps which are
difficult since the users need to search the required news on their
own from a vast data.
 Sentiment analysis to make it more unique.
 Interactive news bot that searches and fetches news for the user.
 The bot understands only a certain fixed keywords.
 Does not have a single full-fledged app.
APPLICATIONS
 Can be used by companies that involve following the latest
news.
 Can be used to make predictions in stock market.
 Can be used by anyone all around the world to get fast and
accurate results.
CONCLUSION
In this project we’ll get to know the basics and build our own
news mining web application using Node-RED / Python Web App
and the IBM Watson Discovery Service. It helps us to know how
things work in the corporate world. We learn how to use slack. It
also makes us aware of the basic applications of AI.
FUTURE SCOPE
 Can be made into a full-fledged single mobile
application.
 Can make the bot a little more advanced.
 The UI can be made more interesting by showing the
latest news and adding more aesthetic front end
designs.
BIBLIOGRAPHY
 https://github.com/
 https://slack.com/intl/en-in/
 https://cloud.ibm.com/

Smart Bridge Internship Report

Uploaded by

Copyright:

Available Formats

Smart Bridge Internship Report

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Smart Bridge Internship Report

Uploaded by

Copyright:

Available Formats

Smart Bridge Internship

AI Powered News Search App

Malla Reddy Institute of Engineering and

All we need to do is give a keyword regarding the required news

It is also not possible to determine the tone or the sentiment of

2.2 Proposed Solution:

By adding sentiment analysis to this application can make the app

1. Build a Server Side Application using Node-RED.

Setting up the development environment

1. Install Git and create a GitHub account. Git installation can be

You might also like