Detection and Prediction of Future Mental Disorder

A Mini Project report on

Detection and Prediction of Future Mental Disorder from Social Media

Data Using Machine Learning
A Dissertation submitted to in partial fulfilment of the academic requirements for
the award of the Degree

Bachelor of Technology
In
INFORMATION TECHNOLOGY

Submitted by
A. Chandra Sena 22H51A1207
Ch. Siri 22H51A1215
Rumaishah 23H55A1208

Under the esteemed guidance of

Dr.K. Venkateshwara Rao
Associate Professor and HOD
Department of IT

Department of Information Technology

CMR COLLEGE OF ENGINEERING & TECHNOLOGY

(UGC AUTONOMOUS)
(NAAC Accredited with ‘A+’ Grade & NBA Accredited)
(Approved by AICTE, Permanently Affiliated to JNTU Hyderabad)
KANDLAKOYA, MEDCHAL ROAD, HYDERABAD – 501401

June 2025

CMR COLLEGE OF ENGINEERING & TECHNOLOGY

 Detection and Prediction of Future Mental Disorder

(UGC AUTONOMOUS)
KANDLAKOYA, MEDCHAL ROAD, HYDERABAD – 501401

DEPARTMENT OF INFORMATION TECHNOLOGY

CERTIFICATE

This is to certify that the Mini Project report entitled " DETECTION AND
PREDICTION OF FUTURE MENTAL DISORDER FROM SOCIAL MEDIA DATA
USING MACHINE LEARNING " being submitted by A.ChandraSena (22H51A1207),
Ch.Siri (22H51A1215), Rumaishah (23H55A1208) in partial fulfilment for the award of
Bachelor of Technology in Information Technology is a record of bonafide work carried
out his/her under my guidance and super vision. This work is original and has not been
submitted to any other University or Institute either in part or in full for the award of any
Degree.

Dr. K. VENKATESWARA RAO Dr. K. VENKATESWARA RAO

Associate Professor and HOD Associate Professor and HOD
Department of IT Department of IT
 Detection and Prediction of Future Mental Disorder

Acknowledgment

With great pleasure we want to take this opportunity to express my heartfelt gratitude to all
the people who helped in making this project work a grand success.
We are grateful to Dr. K. VENKATESWARA RAO, Associate Professor and HOD,
Department of Information Technology for his/her valuable technical suggestions and guidance
during the execution of this project work.
We would like to thank Dr. K. Venkateswara Rao, Associate Professor and Head of the
Department of Information Technology, CMR College of Engineering and Technology, who is the
major driving forces to complete my project work successfully.
We are very grateful to Professor Ghanta Devadas, Dean-Academics, CMR College of
Engineering and Technology, for his constant support and motivation in carrying out the project
work successfully.
We are highly indebted to Dr. A. Sheshu Kumar, Principal, CMR College of Engineering
and Technology, for giving permission to carry out this project in a successful and fruitful way.
We would like to thank the Teaching & Non- Teaching staff of Department of Information
Technology for their co-operation
We express our sincere thanks to Shri. Ch. Gopal Reddy, Secretary, CMR Group of
Institutions, for his continuous care.
Finally, we extend thanks to our parents who stood behind us at different stages of this
Project. We sincerely acknowledge and thank all those who gave support directly and indirectly in
completion of this project work.

A.Chandra Sena 22H51A1207

Ch.Siri 22H51A1215
Rumaishah 23H55A1208
 Detection and Prediction of Future Mental Disorder

DECLARATION

We hereby declare that the project entitled “DETECTION AND PREDICTION OF

FUTURE MENTAL DISORDER FROM SOCIAL MEDIA DATA USING MACHINE
LEARNING” are from the work carried out by using partial fulfillment of the requirements for the
award of B. Tech degree. We have not been submitted to any other University or Institute either in
part or in full for the award of any Degree.

SIGNATURE
A.Chandra Sena 22H51A1207
Ch.Siri 22H51A1215
Rumaishah 23H55A1208
 Detection and Prediction of Future Mental Disorder

ABSTRACT

The increasing use of social media platforms has led to an exponential rise in data related to
individuals' mental health, providing valuable insights for the detection of mental disorders. This
project explores the use of Machine Learning (ML) techniques, particularly Random Forest and
Decision Tree algorithms, to detect potential mental health issues from social media data. By
analyzing the textual content shared by users, the system aims to predict whether a person might be
experiencing a mental health disorder based on their posts and interactions. In this study, we
preprocess the textual data using Natural Language Processing (NLP) techniques, such as
tokenization, stopword removal, and vectorization through the Term Frequency-Inverse Document
Frequency (TF-IDF) method. The preprocessed data is then used to train two widely-used
classification models: Random Forest and Decision Tree. The Random Forest algorithm, an
ensemble learning method, leverages multiple decision trees to improve the accuracy and robustness
of predictions, while the Decision Tree algorithm builds a tree-like model to make predictions based
on feature values. The trained models are evaluated for their classification accuracy, with the
Random Forest model expected to provide better generalization by reducing overfitting compared to
the Decision Tree model. Both models are tested on social media data, with the goal of determining
whether a user’s posts indicate the potential for mental health disorders such as anxiety, depression,
or stress. The results of this study aim to contribute to the development of automated tools for early
mental health detection, which can help in timely intervention and support. In conclusion, the use of
machine learning algorithms such as Random Forest and Decision Tree offers a promising approach
for detecting mental health disorders from social media data, showcasing the potential of AI in the
healthcare domain. This can aid mental health professionals and organizations in providing early
assistance to individuals in need.
 Detection and Prediction of Future Mental Disorder

TABLE OF CONTENTS

CHAPTER NO. TITLE PAGE NO.

Abstract
1 Introduction
1.1 Introduction
1.2 Objective
2 System Analysis
2.1 Functional requirements
2.2 Software requirements
2.3 Hardware requirements
3 Tools & Libraries
3.1 Tools
3.2 Libraries
4 System & Design
4.1 Proposed Solution
4.2 Proposed technique
4.3 System Architecture
5 Implementation
5.1 Implementation
5.2 Source Code
6 Results
7 Conclusion
8 Future Scope
9 References
Detection and Prediction of Future Mental Disorder

LIST OF FIGURES

FIG NO. TITLE OF FIGURE PAGE NO

1 Fig 1 Title
2 Fig 2 Title
Detection and Prediction of Future Mental Disorder

CHAPTER 1

INTRODUCTION
 Detection and Prediction of Future Mental Disorder

1.1 INTRODUCTION

Mental health disorders, such as depression, anxiety, and stress, are increasingly being
recognized as global health concerns. With the rise of social media platforms like Facebook, Twitter,
and Instagram, individuals frequently share personal thoughts, experiences, and emotions, creating a
wealth of data that could potentially offer insights into their mental well-being. By analyzing these
social media posts using machine learning and natural language processing (NLP) techniques, it is
possible to detect early signs of mental health disorders. This project aims to utilize advanced
machine learning models such as Random Forest and Decision Tree to predict mental health
disorders based on text data from social media. The goal is to develop a system that can
automatically identify users who might be at risk, enabling timely interventions and support.

1.2 OBJECTIVE

The main objective of this project is to build an intelligent system that can detect early signs of
mental health disorders by analyzing users' social media content. With the increasing use of
platforms like Twitter, Facebook, and Reddit, people often share their thoughts and emotions online,
providing valuable data for analysis. This project uses machine learning algorithms, specifically
Random Forest and Decision Tree, to classify whether a user might be experiencing mental health
issues such as depression, anxiety, or stress. Natural Language Processing (NLP) techniques are
applied to clean and process the textual data from user posts. These techniques include tokenization,
stopword removal, and TF-IDF vectorization. The processed data is then used to train and test the
models. The goal is to compare the performance of the two algorithms and identify which one
provides more accurate predictions. By doing so, the system aims to support early detection and
intervention. This can help mental health professionals and organizations reach out to individuals in
need at an early stage. Ultimately, the project demonstrates the potential of AI in promoting mental
well-being through automated and data-driven tools.

.
Detection and Prediction of Future Mental Disorder

CHAPTER 2

SYSTEM ANALYSIS
 Detection and Prediction of Future Mental Disorder

2.1 FUNCTIONAL REQUIREMENTS:

The project is designed to detect mental health disorders from social media posts by implementing
key functional components. It must be capable of accepting real-world, anonymized social media
datasets as input, ensuring that all personally identifiable information is removed to maintain user
privacy. The system will utilize Natural Language Processing (NLP) techniques to clean and prepare
the text data. This includes tokenization, stopword removal, lowercasing, and transforming the text
into numerical features using TF-IDF vectorization.

Following preprocessing, the system will implement two classification models: Decision Tree and
Random Forest. The Decision Tree model will offer a transparent, interpretable approach to
classification, while the Random Forest model will enhance accuracy through ensemble learning.
The system must support training and testing of both models using labeled data, allowing evaluation
through metrics such as accuracy, precision, recall, and F1-score. It should also be able to make
predictions on new or unseen posts to determine whether they reflect signs of mental health issues
such as depression, anxiety, or stress. Additionally, the system may include result visualization
features, such as confusion matrices or feature importance charts, and allow exporting results and
performance reports for further analysis or presentation.

2.2HARDWARE REQUIREMENTS:

 Processor - Pentium –IV

 RAM - 4 GB (min)
 Hard Disk - 20 GB
 Key Board - Standard Windows Keyboard
 Mouse - Two or Three Button Mouse
 Monitor - SVGA

2.3 SOFTWARE REQUIREMENTS:

 Operating system : Windows 7 Ultimate

 Coding Language : Python
 Front_End : Python
 Back-End : HTML, Css
 Detection and Prediction of Future Mental Disorder

CHAPTER 3

TOOLS AND LIBRARIES

3.1 TOOLS
 Detection and Prediction of Future Mental Disorder

 Twitter API – For accessing tweets.

 Reddit API – For accessing Reddit posts and comments.
 Scrapy – Web scraping tool for crawling websites.
 BeautifulSoup – For parsing and extracting data from HTML content.
 Regular Expressions (Regex) – For text pattern matching and cleaning.
 TfidfVectorizer – Converts text into numerical features (from scikit-learn).
 Word2Vec, GloVe, FastText – Pre-trained word embedding models.
 VADER Sentiment Analyzer – For sentiment scoring.
 BERT / RoBERTa (via Transformers) – Advanced language understanding models.
 Random Forest & Decision Tree Classifiers – ML algorithms (from scikit-learn).
 XGBoost – Gradient boosting algorithm.
 LightGBM – Lightweight gradient boosting framework.
 Tkinter – For desktop GUI development.
 Flask – For building web applications/APIs.
 Dash – For building analytical web dashboards.

3.2 LIBRARIES

 NLTK – Natural Language Toolkit for text preprocessing.

 SpaCy – NLP library for tokenization, POS tagging, etc.
 scikit-learn – For ML models, TF-IDF, feature selection, and evaluation metrics.
 Hugging Face Transformers – For using and fine-tuning models like BERT and RoBERTa.
 XGBoost – For efficient implementation of gradient boosted trees.
 LightGBM – Fast, scalable gradient boosting framework.
 Regex (re in Python) – Built-in library for regular expressions.
 Pandas – For data manipulation and cleaning.
 NumPy – For numerical operations.
 Matplotlib / Seaborn – For optional data visualization.
 Detection and Prediction of Future Mental Disorder

CHAPTER 4

SYSTEM DESIGN

4.1 PROPOSED SOLUTION:

 Detection and Prediction of Future Mental Disorder

The proposed system seeks to enhance the accuracy and effectiveness of mental health disorder
detection from social media posts by leveraging the power of ensemble learning techniques like
Random Forest and Decision Tree. By applying Natural Language Processing (NLP) techniques to
clean, preprocess, and vectorize the text data, the system will train two distinct models to classify
posts as indicative of potential mental health issues. Random Forest, as an ensemble learning
algorithm, combines the output of several decision trees to improve generalization, while Decision
Trees provide a transparent, easy-to-understand decision-making process. The system will be
evaluated on real-world social media datasets to assess its ability to detect mental health issues with
high accuracy. In addition, the system will be designed with privacy in mind, ensuring that only
necessary, anonymized data is used for training and prediction.

4.2 PROPOSED TECHNIQUE:

The proposed technique involves using Natural Language Processing (NLP) and Machine Learning
to detect potential mental health disorders from social media posts. Text data collected from
platforms like Twitter or Reddit is first preprocessed using NLP methods such as tokenization,
stopword removal, and TF-IDF vectorization. This processed data is then used to train two
classification models: Decision Tree and Random Forest. The Decision Tree offers interpretability by
making decisions based on feature splits, while the Random Forest, an ensemble technique, improves
prediction accuracy by aggregating multiple decision trees to reduce overfitting. The system ensures
user privacy by utilizing only anonymized data and aims to provide accurate and early detection of
mental health issues like anxiety, depression, or stress.

4.3 SYSTEM ARCHITECTURE:

 Detection and Prediction of Future Mental Disorder

ADVANTAGES:

1. Improved Accuracy with Ensemble Learning: By using Random Forest, which combines multiple
decision trees, the system is expected to offer better prediction accuracy and reduced overfitting
compared to traditional single-model approaches.
2. Context-Aware Detection: The system will employ advanced NLP techniques to process social
media text more effectively, capturing contextual nuances such as tone, sentiment, and implied
meaning, which is crucial for detecting mental health disorders.
3. Scalable and Efficient: The system can be scaled to handle large datasets from multiple social
media platforms, offering real-time or near-real-time prediction capabilities, which can aid in
proactive mental health intervention.
 Detection and Prediction of Future Mental Disorder

CHAPTER 5
IMPLEMENTATION

5.1 IMPLEMENTATION:
 Detection and Prediction of Future Mental Disorder

1. Data Collection
 Gather social media data from platforms like Twitter and Reddit using APIs (Twitter API,
Reddit API) or web scraping tools (Scrapy, BeautifulSoup).
 Focus on posts, tweets, or comments that may indicate signs of mental health conditions.
2. Data Cleaning and Preprocessing
 Remove irrelevant content such as URLs, punctuation, and special characters using Regex.
 Convert all text to lowercase.
 Apply tokenization and remove stopwords using NLP libraries like NLTK or SpaCy.
3. Text Vectorization
 Convert cleaned text into numerical format using TF-IDF vectorization with scikit-learn.
 Optionally, explore other embedding techniques (Word2Vec, GloVe) for better context
representation.
4. Feature Engineering
 Perform sentiment analysis using tools like *VADER* or *BERT-based transformers*.
 Extract useful features such as keyword frequency, post length, and mental health-related
terms.
 Apply feature selection techniques (e.g., Mutual Information, RFE) to retain the most
important ones.
5. Model Selection and Training
 Split the dataset into training and testing sets.
 Train two models: *Decision Tree* and *Random Forest* using scikit-learn.
 Tune hyperparameters for optimal performance.
6. Model Evaluation
 Evaluate models using accuracy, precision, recall, and F1-score.
 Compare Decision Tree and Random Forest models to determine which performs better.
 Use tools like confusion matrix and ROC curves for visualization.
7. Prediction and Risk Detection
 Apply the best-performing model to new/unseen text data.
 Predict whether a post indicates potential signs of *depression, **anxiety, or **stress*.
 Optionally assign a *risk score or confidence level* to each prediction.
8. Privacy and Anonymization
 Ensure user privacy by anonymizing all identifiable information before processing.
 Store and process data in a secure environment.

9. User Interface Development

 Detection and Prediction of Future Mental Disorder

 Build a simple UI (using *Tkinter, **Flask, or **Dash*) where users can input social
media text.
 Display prediction results, including detected disorder and risk score.
10. Final Testing and Deployment
 Test the system with real or simulated data to ensure reliability.
 Deploy the model and UI locally or on a web server for demonstration or practical use.

5.2 SOURCE CODE:

 Detection and Prediction of Future Mental Disorder

import tkinter as tk
from tkinter import messagebox, filedialog
import pandas as pd
from sklearn.ensemble import RandomForestClassifier
from sklearn.tree import DecisionTreeClassifier # Import DecisionTreeClassifier
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.model_selection import train_test_split
import torch
import re
import nltk
nltk.download('stopwords')
stop_words = set(nltk.corpus.stopwords.words('english'))
trained_rf_model = None
trained_vectorizer = None
trained_dt_model = None # Variable to store the trained Decision Tree model
df = None # Global dataframe to store CSV data
text_entry = None # Declare text_entry globally
result_text = None # Declare result_text globally
X_train, X_test, y_train, y_test = None, None, None, None
def preprocess_text(text):
if not isinstance(text, str):
return ""
text = re.sub(r"http\S+|www\S+|https\S+", "", text) # Remove URLs
text = re.sub(r"[^A-Za-z\s]", "", text) # Remove non-alphabetical characters
text = text.lower() # Convert to lowercase
text = " ".join([word for word in text.split() if word not in stop_words]) # Remove stopwords
return text
def load_csv():
global df
file_path = filedialog.askopenfilename(filetypes=[("CSV Files", "*.csv")])
 Detection and Prediction of Future Mental Disorder

if file_path:
df = pd.read_csv(file_path)
if 'statement' in df.columns and 'status' in df.columns:
messagebox.showinfo("Success", "CSV loaded successfully.")
else:
messagebox.showerror("Error", "CSV must contain 'statement' and 'status' columns.")
def process_data():
global df, X_train, X_test, y_train, y_test
if df is None:
messagebox.showerror("Error", "Please load a CSV file first.")
return
df['statement'] = df['statement'].fillna('') # Handle missing data
df['processed_text'] = df['statement'].apply(preprocess_text)
vectorizer = TfidfVectorizer(stop_words='english', max_features=1000)
X = vectorizer.fit_transform(df['processed_text'])
y = df['status']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
rf_model = RandomForestClassifier(n_estimators=100, random_state=42)
rf_model.fit(X_train, y_train)
dt_model = DecisionTreeClassifier(random_state=42) # Decision Tree classifier
dt_model.fit(X_train, y_train)
global trained_rf_model, trained_vectorizer, trained_dt_model
trained_rf_model = rf_model
trained_vectorizer = vectorizer
trained_dt_model = dt_model
result_text.delete(1.0, tk.END) # Clear previous output
result_text.insert(tk.END, f"Data split: 80% for training, 20% for testing.\n")
def predict_rf():
if trained_rf_model is None or trained_vectorizer is None:
messagebox.showwarning("Model Error", "Please load data and train the model first.")
 Detection and Prediction of Future Mental Disorder

return
rf_accuracy = trained_rf_model.score(X_test, y_test)
result_text.delete(1.0, tk.END) # Clear previous output
result_text.insert(tk.END, f"RandomForest Accuracy on Test Set: {rf_accuracy * 100:.2f}%\n")
def predict_dt():
if trained_dt_model is None or trained_vectorizer is None:
messagebox.showwarning("Model Error", "Please load data and train the model first.")
return
dt_accuracy = trained_dt_model.score(X_test, y_test)
result_text.delete(1.0, tk.END) # Clear previous output
result_text.insert(tk.END, f"Decision Tree Accuracy on Test Set: {dt_accuracy * 100:.2f}%\n")
def predict_statement():
statement = text_entry.get("1.0", tk.END).strip()
if not statement:
messagebox.showwarning("Input Error", "Please enter a statement for prediction.")
return
processed_statement = preprocess_text(statement)
statement_vector = trained_vectorizer.transform([processed_statement])
rf_prediction = trained_rf_model.predict(statement_vector)[0]
rf_prob = trained_rf_model.predict_proba(statement_vector)[0]
rf_pred_prob = max(rf_prob) * 100
dt_prediction = trained_dt_model.predict(statement_vector)[0]
dt_prob = trained_dt_model.predict_proba(statement_vector)[0]
dt_pred_prob = max(dt_prob) * 100
result_text.delete(1.0, tk.END) # Clear previous output
result_text.insert(tk.END, f"Prediction for Statement: \n")
result_text.insert(tk.END, f"RandomForest Prediction: {rf_prediction}, Probability:
{rf_pred_prob:.2f}%\n")
result_text.insert(tk.END, f"DecisionTree Prediction: {dt_prediction}, Probability:
{dt_pred_prob:.2f}%\n")
def create_gui():
 Detection and Prediction of Future Mental Disorder

global text_entry, result_text

root = tk.Tk()
root.title("Mental Health Disorder Detection")
root.geometry("800x600")
root.config(bg="#9B1B30")
title_label = tk.Label(root,
text="Detection and Prediction of Future Mental Disorder from social media data
using ML, Ensemble learning and LLM",
font=("Arial", 17, "bold"),
bg="white",
fg="black",
padx=20, # Horizontal padding (left and right space)
pady=20) # Vertical padding (top and bottom space)
title_label.pack(pady=20)
button_frame = tk.Frame(root, bg="#9B1B30")
button_frame.pack(side=tk.LEFT, padx=20, pady=20, anchor="n")
load_button = tk.Button(button_frame, text="Load CSV File", font=("Arial", 12),
command=load_csv)
load_button.pack(pady=10, anchor="w")
train_button = tk.Button(button_frame, text="Train and Test", font=("Arial", 12),
command=process_data)
train_button.pack(pady=10, anchor="w")
rf_button = tk.Button(button_frame, text="Run RandomForest Algorithm", font=("Arial", 12),
command=predict_rf)
rf_button.pack(pady=10, anchor="w")
dt_button = tk.Button(button_frame, text="Run Decision Tree Algorithm", font=("Arial", 12),
command=predict_dt)
dt_button.pack(pady=10, anchor="w")
predict_button = tk.Button(button_frame, text="Predict", font=("Arial", 12),
command=predict_statement)
predict_button.pack(pady=10, anchor="w")
text_frame = tk.Frame(root, bg="#9B1B30")
 Detection and Prediction of Future Mental Disorder

text_frame.pack(side=tk.RIGHT, padx=20, pady=20, anchor="n")

text_entry_label = tk.Label(text_frame, text="Enter Statement for Prediction:", font=("Arial", 13),
bg="#9B1B30", fg="white")
text_entry_label.pack()
text_entry = tk.Text(text_frame, height=6, width=50, font=("Arial", 12))
text_entry.pack(pady=10)
result_text = tk.Text(text_frame, height=10, width=50, font=("Arial", 12), bg="#f5f5f5")
result_text.pack(pady=10)
root.mainloop()
if _name_ == "_main_":
create_gui()
 Detection and Prediction of Future Mental Disorder

CHAPTER 6

RESULTS

RESULTS:
Detection and Prediction of Future Mental Disorder
Detection and Prediction of Future Mental Disorder
 Detection and Prediction of Future Mental Disorder

CHAPTER 7

CONCLUSION

CONCLUSION:
 Detection and Prediction of Future Mental Disorder

The detection and prediction of future mental disorders using social media data, combined with
Machine Learning (ML), Ensemble Learning, and Large Language Models (LLMs), represents a
transformative approach in the field of mental health. By leveraging the vast amount of unstructured
data available on social media platforms, machine learning models can identify early warning signs
of mental health issues such as depression, anxiety, and stress. Ensemble learning techniques like
Random Forests and Decision Trees enhance the predictive power of individual models by
combining their strengths and reducing the risk of overfitting. Furthermore, large language models,
which excel in natural language processing tasks, can interpret the subtle and complex language
patterns found in social media content to better assess emotional and psychological states.
This approach offers the potential to create personalized mental health monitoring systems that can
detect issues early, potentially improving treatment outcomes and providing real-time interventions.
While promising, this field still faces challenges, including data privacy concerns, the need for high-
quality labeled datasets, and the interpretability of complex machine learning models. However, as
technology advances and more data becomes available, the capabilities of such systems will continue
to improve, offering a future where mental health disorders can be identified and addressed with
greater precision and efficiency.
Detection and Prediction of Future Mental Disorder

CHAPTER 8

FUTURE SCOPE
 Detection and Prediction of Future Mental Disorder

FUTURE SCOPE:

1. Integration with Wearable Technology: Combining social media data with data from wearable
devices (such as heart rate, sleep patterns, and activity levels) could provide a more holistic view of
an individual's mental health, improving the accuracy of predictions.
2. Real-time Monitoring: With advancements in cloud computing and edge AI, it will be possible to
monitor mental health in real time and provide immediate intervention or support when needed.
3. Cross-platform Analysis: Integrating data from multiple social media platforms (Twitter,
Facebook, Instagram, etc.) and analyzing the interaction between users across these platforms will
provide more accurate mental health insights.
4. Improved Explainability: Future research will focus on improving the interpretability and
transparency of machine learning models, ensuring that predictions are understandable and
actionable for clinicians and patients.
5. Personalized Mental Health Prediction: By utilizing deep learning and personalized data, more
individualized models can be created, offering predictions that account for each person’s unique
mental health profile and life circumstances.
6. Collaboration with Mental Health Experts: Building systems in collaboration with psychologists
and psychiatrists to create models that incorporate clinical expertise could result in more accurate
predictions and better clinical interventions.
7. Ethical Considerations: Future research should also address ethical issues surrounding privacy,
consent, and potential misuse of mental health prediction tools, ensuring that such technologies are
used responsibly.
Detection and Prediction of Future Mental Disorder

CHAPTER 9

REFERENCES
 Detection and Prediction of Future Mental Disorder

REFERENCES:

[1] G. Scott, ‘‘Lifespan (half-life) of social media posts: Update for 2024,’’2024,
doi: 10.13140/RG.2.2.21043.60965.
[2] M. Qian and C. Kong, ‘‘Enabling human-centered machine translationusing
concept-based large language model prompting and translation memory,’’ in Proc.
Int. Conf. Human-Comput. Interact., 2024, pp. 118–134.
[3] I. Frommholz, P. Mayr, G. Cabanac, and S. Verberne, ‘‘Bibliometricenhanced
information retrieval: 14th international BIR workshop (BIR2024),’’ in Proc. Eur.
Conf. Inf. Retr., 2024, pp. 442–446, doi: 10.1007/978-3-031-56069-9_61.
[4] Q. Wang, ‘‘Text memorization: An effective strategy to improve ChineseEFL
learners’ argumentative writing proficiency,’’ Frontiers Psychol.,vol. 14, Apr. 2023,
Art. no. 1126194, doi: 10.3389/fpsyg.2023.1126194.
[5] N. Straková and J. Válek, ‘‘Chatbots as a learning tool: Artificialintelligence in
education,’’RE-Source,pp.245–265,Jan.2024,doi:10.53349/resource.2024.is1 .a1259
[6] R. Thorstad and P. Wolff, ‘‘Predicting future mental illness from socialmedia: A
big-data approach,’’ Behav. Res., vol. 51, pp. 1586–1600,Aug. 2019, doi:
10.3758/s13428-019-01235-z.
[7] [Online]. Available: https://www.reddit.com/dev/api
[8] A. Kumar, A. Sharma, and A. Arora, ‘‘Anxious depression prediction inreal-
time social data,’’ in Proc. Int. Conf. Adv. Eng., Sci., Manag. Technol.,2019, pp. 1–
7.
[9] A. Wongkoblap, M. A. Vadillo, and V. Curcin, ‘‘Predicting social networkusers
with depression from simulated temporal data,’’ in Proc. IEEEEUROCON-18th Int.
Conf. Smart Technol., Jul. 2019, pp. 1–6.
[10] S. Tariq, N. Akhtar, H. Afzal, S. Khalid, M. R. Mufti, S. Hussain, A. Habib,and
G. Ahmad, ‘‘A novel co-training-based approach for the classification of mental
illnesses using social media posts,’’ IEEE Access, vol. 7,pp. 166165–166172, 2019,
doi: 10.1109/ACCESS.2019.2953087.
[11] N. Rezaii, E. Walker, and P. Wolff, ‘‘A machine learning approach
topredicting psychosis using semantic density and latent content analysis,’’Npj
Schizophrenia, vol. 5, no. 1, Jun. 2019.
[12] P. KirindeGamaarachchige and D. Inkpen, ‘‘Multi-task, multi-channel,multi-
input learning for mental illness detection using social media text,’’in Proc. 10th Int.
Workshop Health Text Mining Inf. Anal. (LOUHI),Hong Kong, 2019, pp. 54–64.
[13] A. Trifan, R. Antunes, S. Matos, and J. L. Oliveira, ‘‘Understandingdepression
from psycholinguistic patterns in social media texts,’’ in Proc.Eur. Conf. Inf. Retr.,
vol. 12036, 2020, pp. 402–409.
[14] Z. Jiang, S. I. Levitan, J. Zomick, and J. Hirschberg, ‘‘Detection of
mentalhealth from Reddit via deep contextualized representations,’’ in Proc.
11thInt. Workshop Health Text Mining Inf. Anal., 2020, pp. 147–156.
 Detection and Prediction of Future Mental Disorder

[15] N. S. Alghamdi, H. A. Hosni Mahmoud, A. Abraham, S. A. Alanazi,and L.

García-Hernández, ‘‘Predicting depression symptoms in an Arabicpsychological
forum,’’ IEEE Access, vol. 8, pp. 57317–57334, 2020.
[16] M. L. Birnbaum, R. Norel, A. Van Meter, A. F. Ali, E. Arenare, E. Eyigoz,C.
Agurto, N. Germano, J. M. Kane, and G. A. Cecchi, ‘‘Identifying signalsassociated
with psychiatric illness utilizing language and images posted toFacebook,’’ Npj
Schizophrenia, vol. 6, no. 1, p. 38, Dec. 2020.
[17] R. Chatterjee, R. K. Gupta, and B. Gupta, ‘‘Depression detectionfrom social
media posts using multinomial naive theorem,’’ IOP Conf.Ser., Mater. Sci. Eng.,
vol. 1022, no. 1, 2021, Art. no. 012095, doi:10.1088/1757-899X/1022/1/012095.
[18] L. Ren, H. Lin, B. Xu, S. Zhang, L. Yang, and S. Sun, ‘‘Depressiondetection
on Reddit with an emotion-based attention network: Algorithmdevelopment and
validation,’’ JMIR Med. Informat., vol. 9, no. 7, Jul. 2021,Art. no. e28754.
[19] L. Ansari, S. Ji, Q. Chen, and E. Cambria, ‘‘Ensemble hybrid learning methods
for automated depression detection,’’ IEEE Trans. Computat. Social Syst., vol. 10,
no. 1, pp. 211–219, Feb. 2023, doi:10.1109/TCSS.2022.3154442.
[20] L. Nalini, ‘‘Methods in predictive techniques for mental health status onsocial
networks using machine learning,’’ Int. J. Adv. Res. Sci., Commun.Technol., vol. 2,
no. 1, Jul. 2022.