| dc.contributor.author | Agnihotri, Nisha | |
| dc.contributor.author | Sagar, Shrddha | |
| dc.contributor.author | Singh, Ajay Shankar | |
| dc.date.accessioned | 2024-09-15T06:47:20Z | |
| dc.date.available | 2024-09-15T06:47:20Z | |
| dc.date.issued | 2024-02 | |
| dc.identifier.uri | http://10.10.11.6/handle/1/17984 | |
| dc.description | BIPOLAR DISORDER AND MACHINE LEARNING [1-20]
1.1 Introduction to Bipolar Disorder 1
1.1.1 Bipolar Disorder as a Mental Illness 2
1.1.2 Occurrence of Bipolar Disorder 2
1.1.3 Episodes of Bipolar Disorder 3
1.1.4 Clinical Features of Bipolar Disorder 3
1.1.5 Symptoms of Bipolar Disorder 3
1.1.6 Types of Bipolar Disorder 4
1.1.7 Order and Frequency of Different States of Bipolar Disorder 6
1.1.8 Treatment of Bipolar Disorder 6
1.2 Introduction to Machine Learning and Deep Learning 9
ii
1.2.1 Machine Learning in the Field of Medicine 9
1.2.2 Machine Learning for Diagnosis of Bipolar Disorder 11
1.2.3 Different Models of Machine Learning 12
1.2.4 Supervised Machine Learning 12
1.2.5 Types of Supervised Machine Learning Models 13
1.2.6 Unsupervised Machine learning 14
1.2.7 Types of Unsupervised Models of Machine Learning 15
1.3 Dataset in consideration for the Research 16
1.4 Research Motivation 18
1.5 Research Objective 18
1.6 Organization of the Thesis 19
2. REVIEW OF LITERATURE [21-40]
2.1 Introduction 21
2.2 Review on machine learning techniques to predict bipolar disorder 24
2.2.1 Categories of Mood Disorder 25
2.2.2 Techniques of Machine Learning to Predict Bipolar Disorder 28
2.2.2.1 Analysis Through Background Check and Planning 29
2.2.2.2 Collection or Acquisition of Data 29
2.2.2.3 Data Pre-processing- Feature Selection 30
2.2.2.4 Classifying Data processing Methods 30
2.2.2.5 Evaluation and Prediction Analysis 32
2.2.3 Machine Learning Methodologies for Predicting Bipolar Disorder 32
2.2.4 Conclusion 39
3. PREDICTING THE SYMPTOMS OF BIPOLAR DISORDER USING
MACHINE LEARNING
[41-52]
3.1 Introduction 41
3.2 Categories of Bipolar Disorder 43
3.2.1 Fluctuation of mood episodes 43
iii
3.2.2 Elevation of mood episodes 43
3.2.3 Hypomania with depression 43
3.2.4 Mixed Feature 44
3.2.5 Cycling of mood episodes 44
3.3 Proposed Research work 44
3.3.1 Planning and background Analysis 45
3.3.2 Data Acquisition 45
3.3.3 Data Pre-processing or Data Analysis 45
3.3.4 Implementation and Data Mining Process 45
3.3.5 Performance Evaluation 46
3.4 Literature Review 46
3.5 Research Methodology 47
3.5.1 Dataset 47
3.5.2 Classification 48
3.5.3 Result and Discussion 51
3.6 Conclusion 51
4. BIPOLAR DISORDER: EARLY PREDICTION AND RISK ANALYSIS USING
MACHINE LEARNING
[53-78]
4.1 Introduction 53
4.1.1 Identifying Risk factors of Onset and discourse of bipolar disorder 54
4.1.1.1 Environmental risk factors 55
4.1.1.2 Biological risk factors 55
4.1.1.3 Prodromal Symptoms 55
4.1.1.4 Onset Interval and polarities from initial depression to
Mania
57
4.1.1.5 Conversion from major depression to bipolar disorder 57
4.1.1.6 Bipolar depression in absence of hypomania 57
4.1.2 Related Study 58
iv
4.2 Material and methods 60
4.2.1 Data Preprocessing and Acquisition 62
4.2.1.1 Data Cleaning 64
4.2.1.2 Data Transformation 64
4.2.1.3 Checking for null values 64
4.2.2 Data Analysis 64
4.2.2.1 Training data 65
4.2.2.2 Testing data 65
4.2.3 Data Visualization 65
4.2.4 Classification of Algorithms 67
4.2.5 Performance Evaluation Matrices 68
4.2.5.1 Accuracy 68
4.2.5.2 Precision 69
4.2.5.3 Recall 69
4.2.5.4 F1- Score 69
4.2.5.5 Area under ROC (AUROC) 69
4.2.5.6 Mathews Correlation Coefficient (MCC) 70
4.2.5.7 Cohen’s Kappa 70
4.3
4.2.6 Model Validation
Result
70
72
4.4 Discussion 76
4.5 Conclusion 77
5. IMPLEMENTING MACHINE LEARNING TECHNIQUES TO PREDICT
BIPOLAR DISORDER
[79-98]
5.1 Introduction 79
5.2 Literature Review 81
5.3 Methodology 83
5.3.1 Research Framework 84
v
5.3.2 Characteristics of Participants 85
5.3.2.1 Data Acquisition 86
5.3.2.2 Data Planning and Pre-Processing of Data 86
5.3.2.3 Descriptive Statistics 87
5.3.2.4 Visualization of Data 87
5.3.3 Process Description 92
5.3.3.1 Logistic Regression LR 92
5.3.3.2 Support Vector Mechanism 93
5.3.3.3 K-Nearest Neighbors 93
5.3.3.4 Artificial Neural Networks 93
5.3.3.5 Naïve Bayes 94
5.3.4 Statistical Analysis 94
5.4 Result 94
5.5 Discussion 97
5.6 Conclusion 97
6. ACCURACY ENHANCEMENT USING ARTIFICIAL NEURAL NETWORKS
TO PREDICT BIPOLAR DISORDER
[99-114]
6.1 Introduction 99
6.1.1 Contribution of the Research 100
6.2 Background Review of Related Work 100
6.3 Proposed Model 102
6.3.1
6.3.2
6.3.3
6.3.4
Dataset Acquisition and Preprocessing
Parameterization of Regression Function for Data Analysis
Structure of Artificial Neural Network (ANN) model
Hyperparameters Tuning Parameters for Cross validation
102
103
105
108
6.4 Results and Discussion 109
6.5 Conclusion 113
vi
7. A HYBRID LSVR-LOGISTIC SUPPORT VECTOR REGRESSOR MODEL IS
DESIGNED TO PREDICT BIPOLAR DISORDER
[115-134]
7.1 Introduction 115
7.1.1 Hybrid LSVR Model 116
7.1.2 Contribution of Research 117
7.2 Background Review of Related Work 118
7.3 Logistic Regression & Support Vector Machine Hybrid Model for predicting 122
bipolar disorder
7.3.1 Data Acquisition and Pre-Processing 123
7.3.2 Train the Support Vector Regressor 125
7.3.3 Train the model LR (Logistic Regression) 126
7.3.4 Combine the models 128
7.3.5 Performance Evaluation of the model 129
7.3.6 Deploy the Model 130
7.4 Result 130
7.5 Conclusion | en_US |
| dc.description.abstract | This research can help and guide people having bipolar disorder, their families
and all who wants to understand the basics of this brain disorder, its types, it also focuses
on its treatment and the way to manage with this illness. This is not a substitute for
treatment from a physician or health care practitioner, but it can be used as a basis for
resolving all the queries and discussion related to bipolar disorder. As new drug
treatment and medications are continually being developed, this work will guide people
to be aware of its symptoms and early detection so that the disorder does not turn
hazardous. | en_US |
| dc.language.iso | en | en_US |
| dc.publisher | Galgotias University | en_US |
| dc.subject | Computing Science, Engineering | en_US |
| dc.subject | Phd Thesis | en_US |
| dc.subject | BIPOLAR DISORDER | en_US |
| dc.subject | MACHINE LEARNING | en_US |
| dc.subject | Prediction | en_US |
| dc.title | Detection and Prediction of Bipolar Disorder using Machine Learning Techniques | en_US |
| dc.type | Thesis | en_US |
