Method and System for Presenting Statistical Data in a Natural Language Format

Abstract

A web-based system for presenting statistical analysis in a natural language textual output comprising: a statistics analysis engine, the engine comprising a pre-processing component, an interpretation component, and a natural language generator component, wherein the pre-processing component verifies the data is statistically appropriate for a pre-selected statistical analysis type, the interpretation component interprets the data according to a pre-selected statistical analysis type, and the natural language generator component generates a natural language textual output for the interpreted data.

What is claimed is: 
     
         1 . A computer-implemented method for presenting statistical analysis findings in a natural language textual output comprising:
 providing pre-established plain language descriptive text;   providing a plurality of pre-established statistical analysis types;   selecting at least one of the plurality of pre-established statistical analysis types;   aligning a selected statistical analysis types with a portion of the pre-established plain language descriptive text;   receiving data to be analyzed;   organizing the data;   processing the organized data according to the selected statistical analysis type, wherein if the data is not statistically appropriate for the selected statistical type, replacing the pre-established plain language descriptive text with informative text explaining the data is not appropriate for the selected statistical type;   creating interpreted data by converting the processed data to a pre-determined natural language text;   using one of the pre-established statistical analysis types, determining whether the interpreted data should be processed with a second pre-established statistical analysis type, wherein the second pre-established statistical analysis type is not the previously selected pre-established statistical type;   generating a natural language textual output for the interpreted data according to at least one pre-established rule, the at least one pre-established rule comprising:   dynamically merging at least a portion of the pre-established plain language descriptive text with the natural language text using at least one of the pre-established statistical analysis; and   providing the natural language textual output in a computer file, wherein a computer displays the natural language textual output in a formatted word processing document.   
     
     
         2 . A web-based system for presenting statistical analysis in a natural language textual output, the system comprising:
 a statistics analysis engine communicatively coupled with a web server, wherein the web server receives requests from a client to analyze statistical data, wherein the request includes data to be analyzed and a pre-selected statistical analysis type;   the statistics analysis engine comprising a pre-processing component, an interpretation component, and a natural language generator component;   the pre-processing component configured to analyze the data to verify the data is statistically appropriate for the pre-selected statistical analysis type, wherein if the data is not appropriate for the pre-selected statistical analysis type, the system provides informative text to the client explaining the data is not appropriate for the pre-selected statistical analysis type;   the interpretation component configured to interpret the data using the pre-selected statistical analysis type and convert the data to a pre-determined natural language text, the interpretation component further configured to determine based on the interpreted data a second statistical analysis type to process the data with, wherein the second statistical analysis type is not the same as the pre-selected statistical analysis type;   the natural language generator component configured to generate a natural language textual output for the interpreted data wherein the natural language textual output is a computer file provided to the client wherein when opened the computer file displays the natural language textual output in a formatted word processing document.   
     
     
         3 . The system of  claim 2 , wherein the pre-established statistical analysis type comprises at least one of inferential statistical analyses, non-parametric analysis, data mining analyses (association rules, neural networks, generalized additive models, classification and regression trees, predictive models, CHAID models, cluster analyses, data reduction, classification models), supervised and unsupervised models, time series models, structural equation models, medical informatics, bioinformatics, and fraud detection models. 
     
     
         4 . The system of  claim 2  wherein the data further comprises at least one of numerals and parameters. 
     
     
         5 . The system of  claim 2  wherein the data comprises quantitative data. 
     
     
         6 . The system of  claim 2  wherein the data comprises business data. 
     
     
         7 . The system of  claim 2  further comprising organizing the data to permit the data to be analyzed in accordance with the pre-selected statistical analysis type, thereby providing organized data. 
     
     
         8 . The system of  claim 7  further comprising processing the organized data according to the pre-selected statistical analysis type. 
     
     
         9 . The system of  claim 2  wherein the natural language generator is further configured to generate a natural language textual output according to at least one pre-established rule. 
     
     
         10 . The system of  claim 2  wherein the statistics analysis engine further comprises a text association component, wherein the text association component maps pre-established plain language descriptive text to pre-established statistical analysis types. 
     
     
         11 . The system of  claim 10  wherein the natural language generator is further configured to generate a natural language textual output according to at least one pre-established rule. 
     
     
         12 . The system of  claim 11  wherein the pre-established rule merges at least a portion of the pre-established plain language descriptive text with pre-determined natural language text. 
     
     
         13 . The system of  claim 2  wherein the natural language textual output further comprises a graphic display. 
     
     
         14 . The system of  claim 13  wherein the graphic display is selected from the group consisting of a table, a chart, and a graph.