Deep Analysis for Any Application Domain

Powerful analysis techniques should be applicable to any type of data, ranging from Web analytics to survey data, sales, marketing, insurance, non-profit, engineering, law enforcement, or social data. Whether your data has a few hundred or several hundred thousand records, Data Applied's scalable algorithms can quickly extract valuable knowledge. And because this is what real-world data is made of, Data Applied's analysis techniques handle any mix of numeric, categorical, and missing data. Let us show you what we can do for your data!

Web Analytics

Web Analytics

Example: website clickstream logs

  • Forecast traffic volume given previous trends and variations
  • Visualize page views by keyword, location, and duration
  • Identify which variables most affect bounce behaviors
  • Uncover associations between keywords, locations, and visits
  • Detect visits which are most atypical compared to others
  • Visualize correlations between duration, visitor type, and date
  • Compare average page views, grouping by location, keyword, and date
  • Categorize visits into groups based on visit profiles and dates
  • Project all visits onto a map to visualize similarities between them
  • Apply complex data transformation & preparation to clickstream data
Surveys

Surveys

Example: customer satisfaction survey results

  • Forecast the evolution of customer satisfaction given previous trends
  • Visualize survey answers by respondent age, sex, and location
  • Identify which variables most affect overall customer satisfaction
  • Uncover associations between answers and respondents
  • Detect respondents who provided atypical answers compared to others
  • Visualize correlations between different survey answers
  • Compare survey results, grouping by respondent age, sex, and date
  • Categorize results into groups based on answers and demographics
  • Project all survey answers onto a map to visualize similarities between them
  • Apply complex data transformation & preparation to survey data
Sales

Sales

Example: sales data for multiple products & locations

  • Forecast the evolution of sales amounts given previous trends
  • Visualize sales amounts by product category, store location, and surface
  • Identify which variables most affect the distribution of best selling products
  • Uncover associations between products and customers
  • Detect sales transactions which most differ from similar ones
  • Visualize correlations between total sales, discounts, and surface
  • Compare product sales, grouping by store location and product category
  • Categorize products into groups based on type, location, and amount sold
  • Project all transactions onto a map to visualize similarities between them
  • Apply complex data transformation & preparation to transaction data
Marketing

Marketing

Example: corporate marketing campaign responses

  • Forecast the evolution of customer conversion rates given previous trends
  • Visualize customer conversion by account, contact method, and industry
  • Identify which variables most affect customer conversion
  • Uncover associations between conversion, promotions, and industry
  • Detect customers whose conversion most differ from similar ones
  • Visualize correlations between conversion rate, account size, and loyalty
  • Compare conversion rates, grouping by industry code and contact method
  • Categorize accounts into groups based on conversion, industry, and loyalty
  • Project all responses onto a map to visualize similarities betweeen them
  • Apply complex data transformation & preparation to response data
Insurance

Insurance

Example: automobile insurance reimbursement claims

  • Forecast the evolution of claim rejection rates given previous trends
  • Visualize claim amounts by incident type, location, and processing status
  • Identify which variables most affect claim amounts and rejection rates
  • Uncover associations between claims, incident types, and locations
  • Detect potentially fraudulent claims which most differ from similar ones
  • Visualize correlations between deductibles, reimbursments, and losses
  • Compare claim amounts, grouping by age group, and car model
  • Categorize claims into groups based on car models, drivers, and locations
  • Project all claims onto a map to visualize similarities between them
  • Apply complex data transformation & preparation to incident data
Non-Profit

Non-Profit

Example: amount of cash pledged by college alumni

  • Forecast the evolution of total cash raised given previous trends
  • Visualize amounts pledged by graduation year, field, and GPA
  • Identify which variables most affect pledged amounts and effective payment
  • Uncover associations between pledges, graduation year, and field
  • Detect donors whose pledged amounts most differ from similar ones
  • Visualize correlations between pledged amounts, graduation year, and GPA
  • Compare pledged amounts, grouping by graduation year and field
  • Categorize donors into groups based on pledges, employment, and GPA
  • Project all pledges onto a map to visualize similarities betweeen them
  • Apply complex data transformation & preparation to pledge data
Engineering

Engineering

Example: product manufacturing & quality control data

  • Forecast the evolution of defect rates given previous trends
  • Visualize defects by product, manufacturing line, and manufacture date
  • Identify which variables most affect product defect rates
  • Uncover associations between defects, manufacturing line, and dates
  • Detect products whose characteristics most differ from similar ones
  • Visualize correlations between defects, manufacturing quantities, and dates
  • Compare defects, grouping by manufacturing line and manufacture date
  • Categorize products based on defects, dimensions, and locations
  • Project all products onto a map to visualize similarities betweeen them
  • Apply complex data transformation & preparation to defect data
Law Enforcement

Law Enforcement

Example: neighborood crime & police intervention data

  • Forecast the evolution of crime levels given previous trends
  • Visualize incident counts by neighborood, incident type, and response
  • Identify which variables most affect the distribution of crime types
  • Uncover associations between neighboroods, incidents, and times
  • Detect incidents whose characteristics most differ from similar ones
  • Visualize correlations between incidents, patrol frequency, and dates
  • Compare crime types, grouping by neighborood and incident date
  • Categorize incidents based on crime type, household type, and outcome
  • Project all incidents onto a map to visualize similarities betweeen them
  • Apply complex data transformation & preparation to incident data
Social Sciences

Social Sciences

Example: household immunization health data

  • Forecast the evolution of vaccination levels given previous trends
  • Visualize vaccinations by household income, housing type, and state
  • Identify which variables most affect infection rates and prevention
  • Uncover associations between infection rates, income, and housing types
  • Detect households whose health most differes from similar ones
  • Visualize correlations between infections, income, and square footage
  • Compare vaccinations, grouping by state of residence and income
  • Categorize households into groups based on health, income, and housing
  • Project all households onto a map to visualize similarities betweeen them
  • Apply complex data transformation & preparation to household data