Download as pdf or txt
Download as pdf or txt
You are on page 1of 12

Faculty of Engineering & Technology

Ramaiah University of Applied Sciences

Department Computer Science and Programme B.Tech. in


Engineering CSE/ISE/AI&ML

Semester/Batch 05/2021

Course Code 20AIC301A Course Title Data Mining

Course Leader(s) Dr. Yogesh/ Shona/ Sagar d

Assignment

Register No. 21ETCS002067 Name of Student ABHISHEK KUMAR

Marking Scheme

1.1 Need of data mining in self-driving vehicles 01


Question 1

1.2 Build the data set of 15 instance/candidates/object rows 01

1.3 Determine the logic/ rules to determine the selection 03


process

1.4 Develop a strategy to build an algorithm 03

1.5 Analysis on the acquired rules/logic 02

Question Max Marks 10

Course Marks Tabulation

Assignment Second
First Examiner Remarks Examiner Remarks

Q1

Marks (out of 10 )

Signature of First Examiner Signature of Second Examiner


Assignment

Instructions to students:
1. Maximum marks are 10.
2. The maximum number of pages should be restricted to 5.
3. Submission Date:
4. Submission after the due date is not permitted.
5. Handwritten documents are only accepted for submission
6. Marks will be awarded only to the sections and subsections clearly indicated as per the
problem statement/exercise/question

Assignment
Question 1 (10 Marks)
A Scenario, where driver is driving the 4 wheeler, a self-driving intelligent car which can detect
numerous objects, including humans, their age, profession and character.
An intelligent algorithm is one of the most crucial Techniques used in training self-driving cars. You
have been assigned a task as developer to design a strategy to train your algorithm to manure in
different scenarios
Given a case, where self-driving car is in drive state with faulty machinery leading to eminent crash.
A scenario for drive is as follows:
• Vehicle is on 2 lane path, on the first lane, an individual is stuck with no possibility to walk
away and on the second lane, tree is struck down.
• Algorithm should determine the lane to move forward, where the first lane selection will lead
to the individual’s death and second lane selection leads to driver’s death
Develop a strategy to train the algorithm using, data set details given below:
1. Set the driver state before training the algorithm
Driver states are follows:
Drive_instance Candidate Character Age Profession influence Class
on the
Society
I_1 Driver Good 14 -100 Rich Brat Good Lane 1/
citizen/ No /Bad/No Lane 2
records
(Average)/
Criminal
2. Data Set is as follows:
Drive_instance Candidate Character Age Profession influence Class
on the
Society

I_1 Doctor/Police/Child/ Good 14 - Doctor/Police/student/ Good Lane


citizen/ 100 /Bad/No 1/
Criminal/Rich Criminal/Rich
Kid/Retried IAS No Kid/Retried IAS officer Lane
officer records 2
(Average)/
Criminal

Your report should include:


1.1 Discussion the need of data mining in self-driving vehicles
1.2 Build the data set of 15 instance/candidates/object rows
1.3 Determine the logic/ rules to determine the selection process
1.4 Develop a strategy to build an algorithm to train on developed dataset in 1.2 section (any
classification model) to decide on lane 1 or lane 2 drive path
1.5 Analysis on the acquired rules/logic

You might also like