Method for use of morphological information to cross reference keywords used for information retrieval

Claim 2

A data processing method for building a compressed book file, comprising the steps of: inputting a word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting a document text into said data processor; compiling a vocabulary list of each word in said document text in said data processor; looking up in said morphology dictionary, a lemma for each word from said document text, to identify missing lemmas from said vocabulary list; adding said missing lemmas to said vocabulary list to form an augmented word list; assigning word numbers (WN) to words in said augmented word list; compiling a word number - to - lemma number (WN/LN) list relating word numbers (WN) of words in said augmented word list to their corresponding lemmas, as represented by a lemma number (LN); identifying a lemma number (LN) for each word in said document text by obtaining a word number (WN) from said augmented word list and finding a corresponding LN from said WN/LN list; compiling a lemma number - to - text location list for a location of each word in said document text; substituting a corresponding token for each said word in said document text, forming a compressed tokenized text of said document text; and outputting a compressed book file including said augmented word list, said word number/lemma number list, said compressed tokenized text and said lemma number - to - text location list.

Claim 3

The method of claim 1, which further comprises the method of reading said compressed book file, comprising the steps of: inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; adding said query word to a match list when said query word compares with a word in said augmented word list; searching said word number/lemma number list to identify a lemma number of a lemma of said query word; accessing said lemma of said query word from said augmented word list using said lemma number thereof and adding said lemma of said query word to said match list; accessing a text location value in said document from said lemma number - to - text location list using said lemma number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said query word and said lemma of said query word with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 4

The method of claim 1, which further comprises the method of reading said compressed book file, comprising the steps of: inputting said word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; accessing a lemma of said query word form said word morphology dictionary using said query word; comparing said lemma of said query word with said words in said augmented word list; adding said lemma of said query word to a match list when said lemma of said query word compares with a word having a first word number in said augmented word list; accessing a text location value in said document from said lemma number - to - text location list using said first word number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said lemma of said query word with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 5

The method of claim 1, which further comprises the method of reading said compressed book file, comprising the steps of: inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; adding said query word to a match list when said query word compares with a word in said augmented word list; searching said word number/lemma number list to identify a lemma number of a lemma of said query word; accessing said lemma of said query word from said augmented word list using said lemma number thereof and adding said lemma of said word to said match list; searching said word number/lemma number list to identify a variant word number which has a corresponding lemma number equal to said lemma number of said lemma of said query word; adding said variant word to said match list; accessing a text location value in said document from said lemma number - to - text location list using said lemma number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said query word, said variant word and said lemma of said query word with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 6

The method of claim 1, which further comprises the method of reading said compressed book file, comprising the steps of: inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; adding said query word to a match list when said query word compares with a word in said augmented word list; searching said word number/lemma number list to identify a lemma number of a lemma of said query word; accessing said lemma of said query word from said augmented word list using said lemma number thereof and adding said lemma of said query word to said match list; searching said word number/lemma number list to identify a variant word number for a variant word, which has a corresponding lemma number equal to said lemma number of said lemma of said query word; adding said variant word to said match list; searching said word number/lemma number list to identify a second lemma which has a corresponding variant word with a word number equal to said variant word number; adding said second lemma to said match list; accessing a text location value in said document from said lemma number - to - text location list using said lemma number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said query word, said variant word, said lemma of said query word and said second lemma with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 7

A data processing method for building a compressed book file, comprising the steps of: inputting a word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting a document text into said data processor; compiling a vocabulary list of each word in said document text in said data processor; determining a frequency of occurrence of each word in said document text; looking up in said morphology dictionary, a lemma for each word from said document text, to identify missing lemmas from said vocabulary list; adding said missing lemmas to said vocabulary list to form an augmented word list; assigning word numbers (WN) to words in said augmented word list; compiling a high frequency word pointer list containing entry numbers and corresponding word numbers of high frequency words in said augmented word list; compiling a word number - to - lemma number (WN/LN) list relating word numbers (WN) of words in said augmented word list to their corresponding lemmas, as represented by a lemma number (LN); identifying a lemma number (LN) for each word in said document text by obtaining a word number (WN) from said augmented word list and finding a corresponding LN from said WN/LN list; compiling a lemma number - to - text location list for a location of each word in said document text; substituting a corresponding one of said entry numbers for each said high frequency word in said document text and substituting a corresponding word number for each word of said document text which is not represented in said high frequency pointer list, forming a compressed tokenized text of said document text; and outputting a compressed book file including said high frequency word pointer list, said augmented word list, said word number/lemma number list, said compressed tokenized text and said lemma number - to - text location list.

Claim 8

A data processing method for building a compressed book file, comprising the steps of: inputting a word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting a document text into said data processor; compiling a vocabulary list of each word in said document text in said data processor; determining a frequency of occurrence of each word in said document text; looking up in said morphology dictionary, a lemma for each word from said document text, to identify missing lemmas from said vocabulary list; adding said missing lemmas to said vocabulary list to form an augmented word list; assigning word numbers (WN) to words in said augmented word list; compiling a high frequency word pointer list containing entry numbers and corresponding word numbers of high frequency words in said augmented word list; compiling a word number - to - lemma number (WN/LN) list relating word numbers (WN) of words in said augmented word list to their corresponding lemmas, as represented by a lemma number (LN); identifying a lemma number (LN) for each word in said document text by obtaining a word number (WN) from said augmented word list and finding a corresponding LN from said WN/LN list; compiling a lemma number - to - text location list for a location of each word in said document text; substituting a corresponding one of said entry numbers for each said high frequency word in said document text and substituting a corresponding word number for each word of said document text which is not represented in said high frequency pointer list, forming a compressed tokenized text of said document text; and outputting a compressed book file including said high frequency word pointer list, said augmented word list, said word number/lemma number list, said compressed tokenized text and said lemma number - to - text location list.

Claim 9

The method of claim 6, which further comprises the method of reading said compressed book file, comprising the steps of: inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; adding said query word to a match list when said query word compared with a word in said augmented word list; searching said word number/lemma number list to identify a lemma number of a lemma of said query word; accessing said lemma of said query word from said augmented word list using said lemma number thereof and adding said lemma of said query word to said match list; accessing a text location value in said document from said lemma number - to - text location list using said lemma number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said query word and said lemma of said query word with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 10

The method of claim 6, which further comprises the method of reading said compressed book file, comprising the steps of: inputting said word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; accessing a lemma of said query word from said word morphology dictionary using said query word; comparing said lemma of said query word with said words in said augmented word list; adding said lemma of said query word to a match list when said lemma of said query word compares with a word having a first word number in said augmented word list; accessing a text location value in said document from said lemma number - to - text location list using said first word number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said lemma of said query word with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 11

The method of claim 6, which further comprises the method of reading said compressed book file, comprising the steps of: inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; adding said query word to a match list when said query word compares with a word in said augmented word list; searching said word number/lemma number list to identify a lemma number of a lemma of said query word; accessing said lemma of said query word from said augmented word list using said lemma number thereof and adding said lemma of said query word to said match list; searching said word number/lemma number list to identify a variant word number which has a corresponding lemma number equal to said lemma number of said lemma of said query word; adding said variant word to said match list; accessing a text location value in said document from said lemma number - to - text location list using said lemma number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said query word, said variant word and said lemma of said query word with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 12

The method of claim 6, which further comprises the method of reading said compressed book file, comprising the steps of: inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; adding said query word to a match list when said query word compares with a word in said augmented word list; searching said word number/lemma number list to identify a lemma number of a lemma of said query word; accessing said lemma of said query word from said augmented word list using said lemma number thereof and adding said lemma of said query word to said match list; searching said word number/lemma number list to identify a variant word number for a variant word, which has a corresponding lemma number equal to said lemma number of said lemma of said query word; adding said variant word to said match list; searching said word number/lemma number list to identify a second lemma which has a corresponding variant word with a word number equal to said variant word number; adding said second lemma to said match list; accessing a text location value in said document from said lemma number - to - text location list using said lemma number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said query word, said variant word, said lemma of said query word and said second lemma with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 13

A data processing method for building and reading a compressed book file, comprising the steps of: building a compressed book file comprising the steps of: inputting a word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting a document text into said data processor; compiling a vocabulary list of each word in said document text in said data processor; looking up in said morphology dictionary, a lemma for each word from said document text, to identify missing lemmas from said vocabulary list; adding said missing lemmas to said vocabulary list to form an augmented word list; assigning word numbers (WN) to words in said augmented word list; compiling a word number - to - lemma number (WN/LN) list relating word numbers (WN) of words in said augmented word list to their corresponding lemmas, as represented by a lemma number (LN); identifying a lemma number (LN) for each word in said document text by obtaining a word number (WN) from said augment word list and finding a corresponding LN from said WN/LN list; compiling a lemma number - to - text location list for a location of each word in said document text; substituting a corresponding token for each said word in said document text, forming a compressed tokenized text of said document text; outputting a compressed book file including said augmented word list, said word number/lemma number list, said compressed tokenized text and said lemma number - to - text location list; reading said compressed book file, comprising the steps of: inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; adding said query word to a match list when said query word compares with a word in said augmented word list; searching said word number/lemma number list to identify a lemma number of a lemma of said query word; accessing said lemma of said query word from said augmented word list using said lemma number thereof and adding said lemma of said query word to said match list; accessing a text location value in said document from said lemma number - to - text location list using said lemma number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said query word and said lemma of said query word with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 14

A data processing method for building and reading a compressed book file, comprising the steps of: building a compressed book file comprising the steps of: inputting a word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting a document text into said data processor; compiling a vocabulary list of each word in said document text in said data processor; looking up in said morphology dictionary, a lemma for each word from said document text, to identify missing lemmas from said vocabulary list; adding said missing lemmas to said vocabulary list to form an augmented word list; assigning word numbers (WN) to words in said augmented word list; compiling a word number - to - lemma number (WN/LN) list relating word numbers (WN) of words in said augmented word list to their corresponding lemmas, as represented by a lemma number (LN); identifying a lemma number (LN) for each word in said document text by obtaining a word number (WN) from said augment word list and finding a corresponding LN from said WN/LN list; compiling a lemma number - to - text location list for a location of each word in said document text; substituting a corresponding token for each said word in said document text, forming a compressed tokenized text of said document text; outputting a compressed book file including said augmented word list, said word number/lemma number list, said compressed tokenized text and said lemma number - to - text location list; reading said compressed book file, comprising the steps of: inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; adding said query word to a match list when said query word compares with a word in said augmented word list; searching said word number/lemma number list to identify a lemma number of a lemma of said query word; accessing said lemma of said query word from said augmented word list using said lemma number thereof and adding said lemma of said query word to said match list; accessing a text location value in said document from said lemma number - to - text location list using said lemma number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said query word and said lemma of said query word with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 15

A data processing method for building and reading a compressed book file, comprising the steps of: building a compressed book file comprising the steps of: inputting a word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting a document text into said data processor; compiling a vocabulary list of each word in said document text in said data processor; looking up in said morphology dictionary, a lemma for each word from said document text, to identify missing lemmas from said vocabulary list; adding said missing lemmas to said vocabulary list to form an augmented word list; assigning word numbers (WN) to words in said augmented word list; compiling a word number - to - lemma number (WN/LN) list relating word numbers (WN) of words in said augmented word list to their corresponding lemmas, as represented by a lemma number (LN); identifying a lemma number (LN) for each word in said document text by obtaining a word number (WN) from said augmented word list and finding a corresponding LN from said WN/LN list; compiling a lemma number - to - text location list for a location of each word in said document text; substituting a corresponding token for each said word in said document text, forming a compressed tokenized text of said document text; outputting a compressed book file including said augmented word list, said word number/lemma number list, said compressed tokenized text and said lemma number - to - text location list; reading said compressed book file, comprising the steps of: inputting said word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; accessing a lemma of said query word from said word morphology dictionary using said query word; comparing said lemma of said query word with said words in said augmented word list; adding said lemma of said query word to a match list when said lemma of said query word compares with a word having a first word number in said augmented word list; accessing a text location value in said document from said lemma number - to - text location list using said first word number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said lemma of said query word with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 16

A data processing method for building and reading a compressed book file, comprising the steps of: building a compressed book file comprising the steps of: inputting a word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting a document text into said data processor; compiling a vocabulary list of each word in said document text in said data processor; looking up in said morphology dictionary, a lemma for each word from said document text, to identify missing lemmas from said vocabulary list; adding said missing lemmas to said vocabulary list to form an augmented word list; assigning word numbers (WN) to words in said augmented word list; compiling a word number - to - lemma number (WN/LN) list relating word numbers (WN) of words in said augmented word list to their corresponding lemmas, as represented by a lemma number (LN); identifying a lemma number (LN) for each word in said document text by obtaining a word number (WN) from said augmented word list and finding a corresponding LN from said WN/LN list; compiling a lemma number - to - text location list for a location of each word in said document text; substituting a corresponding token for each said word in said document text, forming a compressed tokenized text of said document text; outputting a compressed book file including said augmented word list, said word number/lemma number list, said compressed tokenized text and said lemma number - to - text location list; reading said compressed book file, comprising the steps of: inputting said word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; accessing a lemma of said query word from said word morphology dictionary using said query word; comparing said lemma of said query word with said words in said augmented word list; adding said lemma of said query word to a match list when said lemma of said query word compares with a word having a first word number in said augmented word list; accessing a text location value in said document from said lemma number - to - text location list using said first word number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said lemma of said query word with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 17

A data processing method for building and reading a compressed book file, comprising the steps of: building a compressed book file comprising the steps of: inputting a word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting a document text into said data processor; compiling a vocabulary list of each word in said document text in said data processor: looking up in said morphology dictionary, a lemma for each word from said document text, to identify missing lemmas from said vocabulary list; adding said missing lemmas to said vocabulary list to form an augmented word list; assigning word numbers (WN) to words in said augmented word list; compiling a word number - to - lemma number (WN/LN) list relating word numbers (WN) of words in said augmented word list to their corresponding lemmas, as represented by a lemma number (LN); identifying a lemma number (LN) for each word in said document text by obtaining a word number (WN) from said augmented word list and finding a corresponding LN from said WN/LN list; compiling a lemma number - to - text location list for a location of each word in said document text; substituting a corresponding token for each said word in said document text, forming a compressed tokenized text of said document text; outputting a compressed book file including said augmented word list, said word number/lemma number list, said compressed tokenized text and said lemma number - to - text location list; reading said compressed book file, comprising the steps of: inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; adding said query word to a match list when said query word compares with a word in said augmented word list; searching said word number/lemma number list to identify a lemma number of a lemma of said query word; accessing said lemma of said query word from said augmented word list using said lemma number thereof and adding said lemma of said query word to said match list; searching said word number/lemma number list to identify a variant word number which has a corresponding lemma number equal to said lemma number of said lemma of said query word; adding said variant word to said match list; accessing a text location value in said document from said lemma number - to - text location list using said lemma number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said query word, said variant word and said lemma of said query word with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 18

A data processing method for building and reading a compressed book file, comprising the steps of: building a compressed book file comprising the steps of: inputting a word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting a document text into said data processor; compiling a vocabulary list of each word in said document text in said data processor: looking up in said morphology dictionary, a lemma for each word from said document text, to identify missing lemmas from said vocabulary list; adding said missing lemmas to said vocabulary list to form an augmented word list; assigning word numbers (WN) to words in said augmented word list; compiling a word number - to - lemma number (WN/LN) list relating word numbers (WN) of words in said augmented word list to their corresponding lemmas, as represented by a lemma number (LN); identifying a lemma number (LN) for each word in said document text by obtaining a word number (WN) from said augmented word list and finding a corresponding LN from said WN/LN list; compiling a lemma number - to - text location list for a location of each word in said document text; substituting a corresponding token for each said word in said document text, forming a compressed tokenized text of said document text; outputting a compressed book file including said augmented word list, said word number/lemma number list, said compressed tokenized text and said lemma number - to - text location list; reading said compressed book file, comprising the steps of: inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; adding said query word to a match list when said query word compares with a word in said augmented word list; searching said word number/lemma number list to identify a lemma number of a lemma of said query word; accessing said lemma of said query word from said augmented word list using said lemma number thereof and adding said lemma of said query word to said match list; searching said word number/lemma number list to identify a variant word number which has a corresponding lemma number equal to said lemma number of said lemma of said query word; adding said variant word to said match list; accessing a text location value in said document from said lemma number - to - text location list using said lemma number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said query word, said variant word and said lemma of said query word with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 19

A data processing method for building and reading a compressed book file, comprising the steps of: building a compressed book file comprising the steps of: inputting a word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting a document text into said data processor; compiling a vocabulary list of each word in said document text in said data processor; looking up in said morphology dictionary, a lemma for each word from said document text, to identify missing lemmas from said vocabulary list; adding said missing lemmas to said vocabulary list to form an augmented word list; assigning word numbers (WN) to words in said augmented word list; compiling a word number - to - lemma number (WN/LN) list relating word numbers (WN) of words in said augmented word list to their corresponding lemmas, as represented by a lemma number (LN); identifying a lemma number (LN) for each word in said document text by obtaining a word number (WN) from said augmented word list and finding a corresponding LN from said WN/LN list; compiling a lemma number - to - text location list for a location of each word in said document text; substituting a corresponding token for each said word in said document text, forming a compressed tokenized text of said document text; outputting a compressed book file including said augmented word list, said word number/lemma number list, said compressed tokenized text and said lemma number - to - text location list; reading said compressed book file, comprising the steps of: inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; adding said query word to a match list when said query word compares with a word in said augmented word list; searching said word number/lemma number list to identify a lemma number of a lemma of said query word; accessing said lemma of said query word from said augmented word list using said lemma number thereof and adding said lemma of said query word to said match list; searching said word number/lemma number list to identify a variant word number for a variant word, which has a corresponding lemma number equal to said lemma number of said lemma of said query word; adding said variant word to said match list; searching said word number/lemma number list to identify a second lemma which has a corresponding variant word with a word number equal to said variant word number; adding said second lemma to said match list; accessing a text location value in said document from said lemma number - to - text location list using said lemma number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said query word, said variant word, said lemma of said query word and said second lemma with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.

Claim 20

A data processing method for building and reading a compressed book file, comprising the steps of: building a compressed book file comprising the steps of: inputting a word morphology dictionary into a data processor, said dictionary including a list of lemmas and linguistic variant words corresponding to said lemmas in said list; inputting a document text into said data processor; compiling a vocabulary list of each word in said document text in said data processor; looking up in said morphology dictionary, a lemma for each word from said document text, to identify missing lemmas from said vocabulary list; adding said missing lemmas to said vocabulary list to form an augmented word list; assigning word numbers (WN) to words in said augmented word list; compiling a word number - to - lemma number (WN/LN) list relating word numbers (WN) of words in said augmented word list to their corresponding lemmas, as represented by a lemma number (LN); identifying a lemma number (LN) for each word in said document text by obtaining a word number (WN) from said augmented word list and finding a corresponding LN from said WN/LN list; compiling a lemma number - to - text location list for a location of each word in said document text; substituting a corresponding token for each said word in said document text, forming a compressed tokenized text of said document text; outputting a compressed book file including said augmented word list, said word number/lemma number list, said compressed tokenized text and said lemma number - to - text location list; reading said compressed book file, comprising the steps of: inputting said compressed book file; inputting a query word; comparing said query word with said words in said augmented word list; adding said query word to a match list when said query word compares with a word in said augmented word list; searching said word number/lemma number list to identify a lemma number of a lemma of said query word; accessing said lemma of said query word from said augmented word list using said lemma number thereof and adding said lemma of said query word to said match list; searching said word number/lemma number list to identify a variant word number for a variant word, which has a corresponding lemma number equal to said lemma number of said lemma of said query word; adding said variant word to said match list; searching said word number/lemma number list to identify a second lemma which has a corresponding variant word with a word number equal to said variant word number; adding said second lemma to said match list; accessing a text location value in said document from said lemma number - to - text location list using said lemma number; accessing a portion of said compressed tokenized text of said document text using said text location value, said portion including a location represented by said text location value; decompressing said portion of said compressed tokenized text, forming a corresponding portion of said document text; comparing said query word, said variant word, said lemma of said query word and said second lemma with words in said corresponding portion of said document text and highlighting words in said corresponding portion which compare.