This document presents a student's final year project on developing a machine learning model to classify hand dominance based on smartphone sensor data. The project aims to collect sensor data from a custom application, preprocess the data, and use the J48 decision tree algorithm to train a model to predict whether a user is right or left handed. Literature on related topics of activity recognition, mobile interface design, and hand gesture recognition is reviewed. The methodology discusses developing an application to collect raw data, preprocessing the data, splitting it into training and testing sets, and evaluating the trained model's performance.
Hand dominant data classification based on smartphone sensitivity
1. Presentation Final Year
Project 1Name : Muhammad Azri Bin Idros
Matric No : 043926
Programme : BACHELOR OF COMPUTER
SCIENCE (NETWORK SECURITY)
WITH HONORS
SUPERVISOR : PROF. MADYA DR. MOHAMAD
AFENDEE MOHAMED
4. 02
There is no
focuses on
the
development
of model that
can predict
whether the
user is right
handed or
left handed
and use it
learn for
themselves
to predict the
result.
01
There is many
intruder out there
that cannot be
traced.
PROBLEM
STATEMENT
There is lack
of machine
learning
application
that have
the hand
dominant
data.
5. PLAN
3
PLAN
2
PLAN
1
OBJECTIVES
To classify
hand
dominant
based on
data that we
get from
smartphone.
Design and
develop
machine
learning
technique to
create a new
model based
on collected
data.
To
implement
the
techniques
of mobile
computing
in the
application
that can test
its
functionality
6. SCOPE
01 02
TARGET USER
Subject will be test
using the
smartphone and we
will get the data
from that.
RESEARCHER
Collect and analyze
the data from the
reading of the
sensors and
classify it based on
sensitivity
smartphone from
7. LIMITATIONS OF WORK
The data that we want to
collect has its own limitation
due to the accuracy.
9. LITERATURE REVIEW
AUTHOR YEAR TITLE SUMMARIZATION
Jennifer R.
Kwapisz,
Gary M.
Weiss,
Samuel A.
Moore
2013 Activity Recognition
using Cell Phone
Accelerometers
A system that uses phone based
accelerometers to perform
activity recognition, a task which
involves identifying the physical
activity a user is performing such
as such as walking, jogging,
climbing stairs, sitting, and
standing. Implement using
system that collected labeled
accelerometer data.
10. LITERATURE REVIEW
AUTHOR YEAR TITLE SUMMARIZATION
Yusof Ahmad 2016 Use of Design Patterns
According to Hand
Dominance in a Mobile
User Interface
User interface (UI) design
patterns for mobile
applications provide a
solution to design
problems and can improve
the usage experience for
users. This research
categorizing the uses of
design patterns according
to users’ hand dominance
in a learning-based mobile
UI.
11. LITERATURE REVIEW
AUTHOR YEAR TITLE SUMMARIZATION
Chris
Bevan,
Danaë
Stanton
Fraser
2016 Different strokes for
different folks?
Revealing the physical
characteristics of
smartphone users from
their swipe gestures
In this paper, it present
initial findings from an
exploration into
whether it is feasible to
infer a single specific
physical characteristic of a
person specifically the
length of their thumb
from the way in which
they perform a common
smartphone interaction
gesture.
12. LITERATURE REVIEW
AUTHOR YEAR TITLE SUMMARIZATION
R.Pradipa,
Ms
S.Kavitha,
Madurai
,Tamil
Nadu
2014 Hand Gesture
Recognition - Analysis of
Various Techniques,
Methods and Their
Algorithms
This survey papers deals
with discussion of various
techniques methods and
algorithms related to the
gesture recognition. The
hand gesture is the most
easy and natural way of
communication.
13. LITERATURE REVIEW
AUTHOR YEAR TITLE SUMMARIZATION
Sunita
Joshi,
Bhuwanes
hwari
Pandey,
Nitin Joshi
2015 Comparative analysis of
Naive Bayes and J48
Classification
Algorithms.
Classification is a data
mining technique based
on machine learning
which is used to
categorize the data item
in a data set into a set of
predefined classes. In this
paper we use two
classification algorithm
J48 and Naïve bayes.
Naïve bayes algorithm is
based on probability and
J48 algorithm is based on
decision tree.
15. Develop an application to
collect data. Collect raw data from
application that we build.
Pre-processed the data.
Split the data into training and
testing.
Interpretation Evaluation
(Knowledge)
FRAMEWORK
17. COLLECT RAW
DATA
Example of raw
data 1
Example of raw
data 2
This is example of raw data. The raw data will
go through to the next process which is pre-
processed data.
18. PRE-PROCESSED
DATA
Data Cleaning: Data
is cleansed through
processes such as
filling in missing
values, smoothing
the noisy data, or
resolving the
inconsistencies in the
data.
Data
Transformation:
Data is
normalized,
aggregated and
generalized.
Data reduction is
the
transformation
of numerical or
alphabetical
digital
information
derived
empirically or
Data wrangling is the
process of
transforming and
mapping data from one
"raw" data form into
another format with the
intent of making it
more appropriate and
valuable for a variety of
21. SELECTED ALGORITHM
What is J48?
J48 J48 decision tree. Then, by applying a
decision tree like J48 on that dataset
would allow us to predict the target
variable of a new dataset record.
23. EXPECTED RESULTS
The project is also
focused on
developing a model
that can determine
whether the user is
right handed or left
handed.
The project
aims to catch
data from
sensitivity of
smartphone.
24. 01
02
03
04
05
Jennifer R. Kwapisz,
Gary M. Weiss,
Samuel A. Moore,
2013,
Activity Recognition
using Cell Phone
Accelerometers
Yusof Ahmad,
2016,
Use of Design
Patterns According
to Hand Dominance
in a Mobile User
Interface
Chris Bevan, Danaë
Stanton Fraser, 2016,
Different strokes for
different folks?
Revealing the physical
characteristics of
smartphone users from
their swipe gestures
Sunita Joshi,
Bhuwaneshwari
Pandey, Nitin
Joshi, 2015,
Comparative
analysis of Naive
Bayes and J48
Classification
R.Pradipa, Ms
S.Kavitha, Madurai
,Tamil Nadu, 2014,
Hand Gesture
Recognition -
Analysis of
Various
Techniques,
REFERENCES