System and Method for Writing Analysis

Abstract

At least one computer readable medium encoded with instructions that, when executed on a computer system, perform a method for determining one or more fluency characteristics of a writer from a piece of their prepared text and providing data to enable characterisation of the writer fluency. The method includes the steps of receiving text data carrying information on the piece of writing to be characterised; enumerating a set of sentences in the piece of writing, wherein each sentence is identified by applying a set of stored rules to the text data; determining a sentence type for each sentence within the piece of text by relating data carrying information on each sentence to stored rules and/or type data; aggregating a data set dependent on the sentence types identified and dependent on a defined aggregation operation; Characterisation data carrying information writer fluency measured dependent on the data set is generated.

1 . A method of determining one or more fluency characteristics of a writer from a piece of their prepared text and outputting feedback to enable characterisation of the writer fluency, the method comprising:
 receiving the piece of text prepared by the writer, wherein the text comprises a plurality of sentences;   determining a sentence style type for each sentence within the text by relating each sentence to rules defining sentence type data, the sentence type data based on a discrete taxonomy of defined sentence style types;   aggregating a data set dependent on;
 the sentence style types identified, and 
 a defined aggregation operation; and 
   generating characterisation data carrying information of writer fluency measured dependent on the data set.   
     
     
         2 . The method of  claim 1 , wherein the discrete taxonomy of defined sentence style types consists in each of:
 Simple Sentence;   Adverb Sentence;   Preposition Sentence;   W-Start Sentence;   Explore the Subject Sentence;   Very Short Sentence;   Em-Dash Sentence;   Ing-Start Sentence;   Ed-Start Sentence;   Serial comma sentence;   The Semi-Colon Sentence;   Question Sentence;   Power Sentence;   Conjunction Sentence;   Colon and Flow Sentence;   Emphatic Ending Sentence;   Developmental flow-on sentence;   Undefined Sentence; and   Incomplete Sentences.   
     
     
         3 . The method of  claim 1 , wherein determining a sentence style type for each sentence comprises:
 applying one or more logic conditions and/or trained machine learning algorithms to the text and characters in a sentence in the text, a match occurring for a particular sentence type when the greatest number of logic conditions are met and/or has a highest probability determined for that sentence type.   
     
     
         4 . The method of  claim 1 , wherein aggregating the data set comprises one or more of:
 summing the sentence style types in the written piece;   determining a ratio between two or more sentence style types in use;   determining a sequence or pattern of two or more sentence style types used in the text;   determining data comparing complete and incomplete sentence style types used in the text;   determining data comparing simple sentence style types to all other sentence style types; and/or   determining the use of precision terms used in one or more sentences.   
     
     
         5 . The method of  claim 1 , wherein characterisation data carrying information of writer fluency is determined by identification of successive use of a particular sentence style type at least three times; and
 generating characterisation data carrying information of writer fluency measured dependent on the data set comprises generating an output indicative of the measure comprising at least one of:   a visual identification of the repetition, and/or   providing a selection of sentence style types for substitution with at least one of the identified repeated sentence types.   
     
     
         6 . The method of  claim 1 , wherein characterisation data carrying information of writer fluency is determined by:
 identifying the sentence style types present and absent in the text;   determining a level of difficulty associated with the absent sentence; and   wherein generating the characterisation data carrying information of writer fluency measured is dependent on the data set comprises a list of the absent sentence style types as list ordered by the level of difficulty of the absent sentence types.   
     
     
         7 . The method of  claim 1 , wherein characterisation data carrying information of writer fluency is determined by:
 identifying a selection of sentence style types;   identifying the order and/or grouping of those sentence style types in the selection;   comparing the identified order and/or grouping to a predetermined selection of superior orders or groupings;   determining at least one absent superior order or grouping; and   wherein generating the characterisation data carrying information of writer fluency measured is dependent on the data set comprises at least one absent superior orders or groupings.   
     
     
         8 . The method of  claim 1 , wherein characterisation data carrying information of writer fluency is determined by:
 identifying a selection of sentence style types;   identifying the frequency of those sentence style types;   comparing the frequency to at least one predetermined superior or reference frequency; and   wherein generating the characterisation data carrying information of writer fluency measured is dependent on the data set comprises that frequency.   
     
     
         9 . The method of  claim 1 , wherein characterisation data carrying information of writer fluency is determined by:
 identifying a selection of sentence style types used by the writer;   identifying the sentence type and order of those sentence style types;   comparing the type and order to the type and order of a reference text; and   wherein generating the characterisation data carrying information of writer fluency measured is dependent on the data set comprises a comparison of the type and order of the text and reference text.   
     
     
         10 . The method of  claim 1 , wherein characterisation data carrying information of writer fluency is determined by:
 identifying a measure of either the rate of precision terms, or the density of precision terms, or both; and   wherein generating the characterisation data carrying information of writer fluency measured is dependent on the data set comprises the rate of precision terms and/or the density of precision terms.   
     
     
         11 . The method of  claim 1 , wherein generating the characterisation data carrying information of writer fluency measured is dependent on the data set comprises one or more of:
 display of a fluency chart;   identification of a particular sentence style type that has been used successively three or more times; and/or   identification of one or more sentence style types that are not used, or least used in the text.   
     
     
         12 . The method of  claim 1 , wherein characterisation data carrying information of writer fluency is determined by:
 identifying each sentence as a complete sentence or incomplete sentence, the incomplete sentence being the incomplete sentence style type, and the complete sentence is any other sentence style type; determining a ration of complete sentences to incomplete sentences; and   wherein generating the characterisation data carrying information of writer fluency measured is dependent on the data set comprises the ratio of complete to incomplete sentences.   
     
     
         13 . The method of  claim 1 , wherein characterisation data carrying information of writer fluency is determined by:
 identifying the sequence of sentence style types deployed in the text;   determining the use of the same sentence style types above a threshold repetition frequency; and   generating characterisation data carrying information of writer fluency measured dependent on the data set comprises data indicative of sequential use of same sentence style types.   
     
     
         14 . The method of  claim 13 , wherein the threshold for each sentence style type is:
 Simple Sentence; 4   Adverb Sentence; 2   Preposition Sentence; 2   W-Start Sentence; 2   Explore the Subject Sentence; 2   Very Short Sentence; 3   Em-Dash Sentence; 2   Ing-Start Sentence; 1   Ed-Start Sentence; 2   Serial comma (Red, White, and Blue); 1   The Semi-Colon Sentence; 1   Question Sentence; 2   Power Sentence; 2   Conjunction Sentence; 1   Colon and Flow Sentence; 1   Emphatic Ending Sentence; 1   Developmental flow-on sentence; 1   Undefined Sentence; and 2   Incomplete Sentences. 1   
     
     
         15 . The method of  claim 1 , wherein characterisation data carrying information of writer fluency is determined by:
 identifying one or more sentences absent of precision terms from a predetermined list of precision terms; and   generating characterisation data carrying information of writer fluency measured dependent on the data set comprises data indicative of the sentences absent of precision terms.   
     
     
         16 . The method of  claim 1 , wherein characterisation data carrying information of writer fluency is determined by:
 aggregating a ratio representing sentences with precisions terms and sentences absent of precision terms;   comparing the ratio to a threshold; and   generating characterisation data carrying information of writer fluency measured dependent on the data set comprises data indicative of the ratio being below the threshold.   
     
     
         17 .- 19 . (canceled) 
     
     
         20 . The method of  claim 1  further comprising output of feedback to enable characterisation of the writer precision, the method comprising:
 determining each sentence in the piece of written text; 
 identifying, within each determined sentence, one or more words indicative of a precision term; 
 outputting data indicative of precision term use in the text and the precision characteristics of the writer. 
 
     
     
         21 . The method of  claim 20 , wherein the precision terms comprise at least one of actual names, dates, statistics, amounts, events, places, people, things, and including subject terminology. 
     
     
         22 . The method of  claim 20 , further comprising calculating the rate of precision term use as the number of sentences containing precision elements divided by the number of sentences identified in the text; and
 outputting data indicative of precision term use in the text and the precision characteristics of the writer comprises scoring the writer based on the calculated rate of precision term use.   
     
     
         23 . The method of  claim 20 , further comprising calculating the number of precision terms used per the number of words in the text, the calculation indicative of the density of precision term use; and
 outputting data indicative of precision term use in the text and the precision characteristics of the writer comprises scoring the writer based on the calculated density of precision term use.   
     
     
         24 . The method of  claim 20 , further comprising determining one or more sentences where precision terms are absent; then outputting data indicative of the sentences where precision terms are absent. 
     
     
         25 . The method of  claim 20 , further comprising calculating a ratio of sentences with precision terms to sentences without precision terms; and
 outputting data indicative of precision term use in the text and the precision characteristics of the writer comprises scoring the writer based on the ratio, and/or generating a warning when the ratio is below the threshold, the threshold dependent on one or more characteristics of the writer, including age and/or writing skill level and/or previous work by the writer.   
     
     
         26 . The method  claim 20 , further comprising:
 determining whether a sentence has a word count exceeding a desired threshold, and outputting feedback to the writer when that threshold is exceed; and   wherein the threshold is dependent on whether or not at least one precision term has been identified.   
     
     
         27 .- 28 . (canceled) 
     
     
         29 . A system configured to determine one or more fluency characteristics of a writer from a piece of their prepared text and output feedback to enable characterisation of the writer fluency, the system comprising:
 a storage component configured to store processor executable instructions; and   a processor configured to:
 receive the piece of text prepared by the writer, wherein the text comprises a plurality of sentences; 
 determine a sentence style type for each sentence within the text by relating each sentence to stored rules defining sentence type data, the sentence type data based on a discrete taxonomy of defined sentence style types; 
 aggregate a data set dependent on:
 the sentence types identified, and 
 a defined aggregation operation; and 
 
 generate characterisation data carrying information of writer fluency measured dependent on the data set. 
   
     
     
         30 . At least one computer readable medium encoded with instructions that, when executed on a computer system, perform a method for determining one or more fluency characteristics of a writer from a piece of their prepared text and providing data to enable characterisation of the writer fluency, the method comprising acts of:
 receiving the piece of text prepared by the writer, wherein the text comprises a plurality of sentences;   determining a sentence style type for each sentence within the text by relating each sentence to stored rules defining sentence type data, the sentence type data based on a discrete taxonomy of defined sentence style types;   aggregating a data set dependent on:
 the sentence types identified, and 
 a defined aggregation operation; and 
   generating characterisation data carrying information of writer fluency measured dependent on the data set.