Method and System for Presenting Statistical Data in a Natural Language Format
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.
This application is a continuation of application Ser. No. 15/213,464, filed Jul. 19, 2016, entitled “Method and System for Presenting Statistical Data in a Natural Language Format,” which is a continuation-in-part of application Ser. No. 14/040,029, filed Sep. 27, 2013, entitled “Method and System for Presenting Statistical Data in a Natural Language Format,” the contents of which are fully incorporated by reference herein for all purposes.
TECHNICAL FIELDThis disclosure relates to a method for processing and interpreting statistical analyses and presenting the results of the analyses as plain language text. More particularly, this disclosure relates to the interpretation of quantitative data on a battery of surveys thereby presenting a profile.
BACKGROUND OF THE INVENTIONStatistical analysis is vital in almost every industry. But connecting mathematical values and results to their natural language meaning has troubled generations. While experts are able to understand the meaning of the statistical analysis, those with less expertise are not.
Thus, there exists a need in the industry to overcome these problems (and others) and provide a method and system for presenting statistical analysis in a natural language textual output. Additionally, there exists a need to interpret the statistical analysis, and convert the interpreted statistical analysis into natural language prose.
SUMMARY OF THE INVENTIONAccording to one embodiment, this disclosure provides a computer-implemented method for presenting statistical analysis in a natural language textual output, the computer-implemented method employing a processor and comprising receiving data to be analyzed by the processor and processing the data according to at least one of a plurality of pre-established statistical analysis types, thereby providing processed data. The method may further comprise interpreting the processed data by converting the processed data to a pre-determined natural language text, thereby providing interpreted data, and generating a natural language textual output for the interpreted data according to at least one pre-established rule for converting the interpreted data to a natural language textual output. The computer-implemented method may be configured to function in a dynamic real-time mode.
The disclosed method has several important advantages. For example, the disclosed method permits the rapid interpretation of complex statistical data by a layperson.
A further possible advantage of the disclosed method includes enabling a financial investor to rapidly interpret changes to a stock value and decide whether to buy or sell a particular stock.
Various embodiments of the present disclosure may have none, some, or all of these advantages. Other technical advantages of the present disclosure may also be readily apparent to one skilled in the art.
For a more complete understanding of the present disclosure and its advantages, reference is now made to the following descriptions, taken in conjunction with the accompanying drawings, in which:
Similar reference numerals refer to similar parts or steps throughout the several views of the drawings.
DETAILED DESCRIPTION OF THE EMBODIMENTSThe present disclosure relates to a method for analyzing statistical data and presenting the analysis in a textual or prose format (i.e., common language). The method will use stored rules for determining the interpretation of the statistical data and the text of the output. The method can function in conjunction with a computer processor in a dynamic, real-time mode. One possible example is the interpretation of quantitative data on a stock value over a time frame and providing a textual report to less expert recipients, including average stock holders or decision makers. The various components of the present invention, and the manner in which they interrelate, are described in greater detail hereinafter.
Initially with reference to
One embodiment of the present invention further comprises selecting at least one of the plurality of pre-established statistical analysis types 64, aligning at least one of the plurality of pre-established statistical analysis types with at least a portion of the pre-established plain language descriptive text 68, receiving data to be analyzed by the processor 72, wherein the data comprises at least one of numerals and parameters.
Another embodiment of the present invention further comprises organizing the data to permit the data to be analyzed in accordance with the selected pre-established statistical analysis type 80, thereby providing organized data, and processing the organized data according to the selected pre-established statistical analysis type 88, wherein the processing is performed by a web-based open source statistical analysis tool, for providing processed data.
Yet another embodiment of the invention further comprises interpreting the processed data by converting the processed data to pre-determined natural language text 100, thereby providing interpreted data.
Another embodiment of the invention further comprises generating a natural language textual output 108 for the interpreted data according to at least one pre-established rule, the at least one pre-established rule comprising merging at least a portion of the pre-established plain language descriptive text with the natural language text, wherein the output further comprises at least one graphic display, and wherein the graphic display is selected from the group consisting of a table, a chart, and a graph, and providing the natural language textual output in a computer file 112, wherein the computer file displays the natural language textual output.
Turning to
In a preferred embodiment, the statistics engine 202 uses an open source statistical analysis tool such as R. The statistics engine 202 is preferably configured to process data to be processed. Thus, the statistics engine 202 preferably receives raw or pre-formatted data and performs statistical analysis on that data. The data (both before and after processing) may be stored or persisted in one or more databases 208. In one embodiment, the database 208 is a MySQL database.
The interpretation engine 204 receives the statistical analysis data from the statistics engine 204 and interprets that data according to pre-configured rules. These rules may also be stored in one or more databases 208, or may be hard-coded into the interpretation engine 204. The interpretation rules may be stored separately from the data itself.
The recommendation engine 206 receives the interpreted data and generates natural language textual output based upon the interpreted data. This output is preferably recommendations to the reader based upon the statistical analysis and interpretation performed by the statistics engine 202 and interpretation engine 204.
In one preferred embodiment, the recommendation engine 206 generates a word processing file, such as a MICROSOFT WORD file. This word processing file includes natural language prose providing analysis and recommendations based on the data processed by the system 200.
The below pseudo-code demonstrates a preferred embodiment for verifying whether or not the data is appropriate for the pre-selected analysis type:
As shown above, for a statistical analysis type “Type A”, the method may determine whether the data provided is sufficient. For this example, certain informative text is provided (ie. “With a sufficiently large sample size (n>100), deviations from normality will have little effect on the results”) if the data provided is not appropriate for the pre-selected analysis.
Returning to
The following pseudo-code shows a preferred embodiment for determining whether the interpreted data would benefit from another statistical analysis type:
As shown above, based on the input data, particular informative text is provided back to the client. After “test.a” is performed, the results of “test.a” are reviewed and based on those results, (i.e. “test.a<100”) the system determines another test (i.e. test.b) should be performed.
It should be appreciated by one skilled in the art that the methods described herein can be configured to function in a dynamic, real-time mode.
It will be understood by those of skill in the art that flowcharts and block diagrams herein described may illustrate architecture, algorithms, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments. Therefore, it will be understood that each block in the flowchart or block diagram may represent a module, segment, or portion of code, which comprises one or more executable computer program instructions for implementing the specified logical function or functions. Further, some implementations may include the functions in the blocks occurring out of the order as herein presented. By way of non-limiting example, two blocks shown in succession may be executed substantially concurrently, or the blocks may at times be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and flowcharts, and combinations of blocks in the block diagram and flowchart illustrations, may be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer program instructions.
The method 100, 300 and system 200 may preferably be implemented in a computing system, which can include a personal computer, a workstation, a network computer, a hand held computer, or any other computing system. Further, the system 200 may be written as a software program in any appropriate computer language.
The system 200 includes one or more processing devices, which may be any computer processing unit, and could be a single central processing unit, or a number of processing units configured to operate either in sequence or in parallel. The processing device can be configured to execute software processes which implement the steps disclosed herein. The system may also include a memory capable of storing the steps necessary for a processing device to implement the steps disclosed herein. This memory could be in the form of memory resident within the processing device or in the form of standalone memory coupled to the processing unit via a communication path, such as a bus or a network.
Although this disclosure has been described in terms of certain embodiments and generally associated methods, alterations and permutations of these embodiments and methods will be apparent to those skilled in the art. Accordingly, the above description of example embodiments does not constrain this disclosure. Other changes, substitutions, and alterations are also possible without departing from the spirit and scope of this disclosure.
Claims
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.
Type: Application
Filed: Sep 14, 2017
Publication Date: Mar 8, 2018
Inventors: James A. Lani (Clearwater, FL), Melissa Moran (Oldsmar, FL)
Application Number: 15/704,037