Congratulations!

Cleaning Data in Java

Dennis Lee

Software Engineer

Journey through data cleaning

// From assessing quality
DescriptiveStatistics stats = new DescriptiveStatistics(values);
System.out.println("Mean: " + stats.getMean());


// To transforming strings
String clean = text.trim().toLowerCase();


// To validating ranges
boolean isValid = Range.between(0, 100).contains(value);


// To cleaning tables
Table cleaned = data.where(sales.isGreaterThan(1000));

Assessing data quality

Bad data leads to bad decisions
Assess data quality: check for outliers, missing values, wrong types

// Check statistics
double mean = stats.getMean();


// Detect nulls
boolean isMissing = Optional.ofNullable(value).isEmpty();


// Verify types
boolean isValidDate = LocalDate.parse(dateStr, formatter);

Transforming data consistently

Inconsistent formats prevent accurate analysis
Clean strings; standardize categories and dates

// Normalize strings (e.g., remove non-letters)
String cleaned = dirtyName.replaceAll("[^a-zA-Z\\s]", "");


// Standardize categories
String standard = categoryMap.getOrDefault(value, "Other");


// Convert dates
LocalDateTime datetime = LocalDateTime.parse(dateStr, formatter);

Validating data integrity

Stop bad data before it enters our system
Check for valid ranges and formats against business rules

// Check ranges
boolean inRange = Range.between(min, max).contains(value);


// Validate patterns
boolean matches = Pattern.matches(pattern, text);


// Enforce constraints
Set<ConstraintViolation<Data>> violations = validator.validate(data);

Cleaning tabular data

for (String colName : data.columnNames()) {
    // Check missing values 
    System.out.println(data.column(colName).countMissing());
}


StringColumn cleaned = names.map(String::toLowerCase)   // Clean text
        .setName("Clean_Names");                        // Name new column


Table summary = data
        .where(sales.isGreaterThan(1000))               // Filter rows
        .summarize("Sales", mean)                       // Aggregate data
        .by("Category");                                // Group results

Resources

Article: Regular Expressions Cheat Sheet
Course: Optimizing Code in Java
Course: Introduction to Testing in Java

Ready to clean

Cleaning Data in Java