PDF Bank Statement Parser

A Java-based PDF bank statement parsing and transaction extraction project built using PDFBox and Tabula.

This project extracts tabular transaction data from digitally generated bank e-statements, normalizes it into structured Transaction objects, and generates validated financial summaries including deposits, withdrawals, and closing balances.

The project was developed as a standalone parser and data normalization pipeline focused on handling semi-structured financial PDF documents.

---

Features

• Extracts tabular transaction data from PDF bank statements
• Uses Apache PDFBox for PDF loading and document processing
• Uses Tabula for table extraction
• Converts extracted rows into normalized Transaction domain objects
• Handles split balance columns caused by PDF formatting inconsistencies
• Skips repeated table headers and summary rows
• Preserves empty columns for consistent schema handling
• Normalizes numeric values for easier parsing and processing
• Processes multiple PDF statements in batch mode
• Generates statement summaries including:
  • total deposits
  • total withdrawals
  • closing balance
• Validates extracted totals against original PDF statements

---

Architecture Flow

PDF Statement
    ↓
Table Extraction (Tabula)
    ↓
Row Normalization
    ↓
Transaction Object Mapping
    ↓
Statement Summary Generation

---

Tech Stack

• Java
• Maven
• Apache PDFBox
• Tabula

---

Project Structure

src/
 └── main/
     └── java/
         └── org/example/
             ├── Main.java
             ├── PdfParser.java
             └── Transaction.java

---

Transaction Schema

Each transaction is normalized into the following structure:

Field	Description
Date	Transaction date
Value Date	Settlement/value date
Description	Transaction description
Cheque	Cheque number (if available)
Deposit	Deposit amount
Withdrawal	Withdrawal amount
Balance	Account balance after transaction

---

Example Output

fileName                deposit     withdrawal     balance

estatement-2025_01.pdf  34767.00    37999.51       24908.55
estatement-2025_02.pdf  31289.70    29663.87       30856.83
estatement-2025_03.pdf  29279.00    12948.20       51514.63

---

Dependencies

Add the following dependencies to pom.xml:

<dependencies>

    <dependency>
        <groupId>org.apache.pdfbox</groupId>
        <artifactId>pdfbox</artifactId>
        <version>3.0.7</version>
    </dependency>

    <dependency>
        <groupId>technology.tabula</groupId>
        <artifactId>tabula</artifactId>
        <version>1.0.5</version>
    </dependency>

</dependencies>

---

Running the Project

Clone the repository

git clone https://github.com/Aum-kar/java-pdf-parser.git

Open the project

Open the project using IntelliJ IDEA or another Java IDE.

Add PDF statements

Place PDF bank statement files inside a local directory.

Update the directory path in Main.java:

File directory = new File("path/to/pdf/folder/");

Run the application

Execute Main.java.

---

Current Limitations

• Optimized for digitally generated bank statements
• Some long descriptions may be partially truncated depending on PDF layout
• Does not support scanned/image-based PDFs
• Parsing logic may require adjustments for different bank formats
• Footer/legal disclaimer text may occasionally be extracted and filtered during parsing

---

Project Status

This project is considered feature-complete in its current form and serves as a standalone PDF parsing and transaction extraction pipeline.

Future development, database integration, and web-based workflows will be implemented separately in a new project.

---

License

This project was developed for educational and personal use.