Lab Assignment 4 Confusion Matrix for Image Classfication & Google Conversational App - rashmitripathi/Big_Data_Analytics_And_Apps GitHub Wiki

1 Write a spark program for the following Machine Learning Tasks.

Create your own dataset for Image Classification Problem. Use the workflow as discussed in the Tutorial 4 Session using Decision Tree Algorithm. Report the accuracy and confusion matrix obtained.Include a brief description of your dataset and purpose behind image classification problem

As part of this program, I have selected the following broad categories of various images for user liking on house interior decoration.

Scenario:

User is shown multiple home interior decoration ideas and based on their liking, we can select the images which are more likely to cater the particular set of users.Users can be categorized as simple, trendy, modern, classic.

Image categories used are:

val IMAGE_CATEGORIES = List("sofa","chair", "table", "wall","plant","guitar", "lamp", "basket","clock","mobile")

Images used are for testing:

Images used for training:

Confusion matrix and Accuracy obtained:

2. Google Conversion API:

Build a simple application to have a conversion about the summary you had generated about your video in Lab Assignment

As part of this, I have used API.AI and deployed the JAR on Heroku. Then the same webhook URL has been used at API.ai.

Heroku Deployment procedure as shown below:

Create Heroku app at command prompt.

Checked deployment at end site and then webhook URL

Associated webhook URL to API.AI

Intents described are:

Intents Source code

Entity described is:

EntitySource code

Conversation Screenshots

References:

Class Tutorial