Systems and methods for generating dialog trees
A dialog tree generation system is provided, including a processor, and a memory storing instructions that, when executed by the processor, cause the system to receive documents, parse the documents into raw blocks, extract visual design elements from the raw blocks, generate a content structure from the extracted visual design elements, generate at least a dialog decision tree based on the extracted content structure, the dialog decision tree comprising a plurality of nodes organized into a hierarchy, and output the dialog decision tree.
1 . A dialog tree generation system comprising:
a processor, and a memory storing instructions that, when executed by the processor, cause the dialog tree generation system to:
receive documents;
parse the documents into raw blocks;
extract visual design elements from the raw blocks, the extracted visual design elements including style and geometry properties of the raw blocks;
clustering the raw blocks based on the extracted visual design elements of the raw blocks;
arrange the clustered raw blocks into a hierarchy;
generate a content structure from the clustered raw blocks arranged into the hierarchy;
generate at least a dialog decision tree based on the content structure, the dialog decision tree comprising a plurality of nodes organized into the hierarchy, and levels of the content structure correspond to levels of the dialog decision tree, wherein the dialog decision tree includes a binary type node branch from a node to two topics and an entity based node branching from a node to more than two topics;
output the dialog decision tree;
execute a virtual assistant to execute the dialog decision tree;
receive a user input;
process the user input with the dialog decision tree; and
generate an output based on the executed dialog decision tree, wherein
the style and geometry properties include at least one of font and typographical properties, paragraph alignment, line lengths, column widths, or graphic shapes,
the outputted dialog decision tree is executed by the virtual assistant to thereby present an interactive chat dialog to a user that proceeds according to the dialog decision tree and includes a binary type dialog based on the binary type node branch and an entity based dialog based on the entity based node branching.
2 . The dialog tree generation system of claim 1 , wherein the raw blocks comprise raw text blocks, raw image blocks, and raw shape blocks;
the style and geometry properties of the raw text blocks include color spaces and text dimensions.
3 . The dialog tree generation system of claim 2 , wherein
the style and geometry properties of the raw shape blocks include vector properties; and
the style and geometry properties of the raw image blocks include positions, dimensions, encodings, and color spaces.
4 . The dialog tree generation system of claim 1 , wherein the extracted visual design elements comprise a page layout including at least one of paragraphs, lists, columns, charts, tables, captions, table of contents, or indices.
5 . The dialog tree generation system of claim 4 , wherein the tables are determined by clustering lines of the raw blocks into separate bins and detecting boundaries of the bins or segmenting table columns/rows based on alignment for borderless tables.
6 . The dialog tree generation system of claim 1 , wherein the content structure includes titles, headers, footers, headings, and heading trees.
7 . The dialog tree generation system of claim 1 , wherein the content structure is a hierarchical graph data structure in which nodes are organized in a traversable tree.
8 . The dialog tree generation system of claim 1 , wherein the content structure is annotated with action entities defining actions and generic entities defining key phrases.
9 . The dialog tree generation system of claim 1 , wherein the font and typographical properties include at least one of typeface, letterforms, font size, font family, style, color, orientation, length, contrast, position, spacing, or dimension.
10 . A dialog tree generation method comprising:
receiving documents;
parsing the documents into raw blocks;
extracting visual design elements from the raw blocks, the extracted visual design elements including style and geometry properties of the raw blocks;
clustering the raw blocks based on the extracted visual design elements of the raw blocks;
arranging the clustered raw blocks into a hierarchy;
generating a content structure from the clustered raw blocks arranged into the hierarchy;
generating at least a dialog decision tree based on the content structure, the dialog decision tree comprising a plurality of nodes organized into the hierarchy, and levels of the content structure correspond to levels of the dialog decision tree, wherein the dialog decision tree includes a binary type node branch from a node to two topics and an entity based node branching from a node to more than two topics;
outputting the dialog decision tree;
executing a virtual assistant to execute the dialog decision tree;
receiving a user input;
processing the user input with the dialog decision tree; and
generating an output based on the executed dialog decision tree, wherein
the style and geometry properties include at least one of font and typographical properties, paragraph alignment, line lengths, column widths, or graphic shapes, and
the outputted dialog decision tree is executed by the virtual assistant to thereby present an interactive chat dialog to a user that proceeds according to the dialog decision tree and includes binary type dialog based on the binary type node branch and an entity based dialog based on the entity based node branching.
11 . The dialog tree generation method of claim 10 , wherein
the raw blocks comprise raw text blocks, raw image blocks, and raw shape blocks;
the style and geometry properties of the raw text blocks include color spaces and text dimensions;
the style and geometry properties of the raw shape blocks include vector properties; and
the style and geometry properties of the raw image blocks include positions, dimensions, encodings, and color spaces.
12 . The dialog tree generation method of claim 10 , wherein the extracted visual design elements include a page layout.
13 . The dialog tree generation method of claim 12 , wherein the page layout includes at least one of paragraphs, lists, columns, charts, tables, captions, table of contents, or indices.
14 . The dialog tree generation method of claim 13 , wherein the tables are determined by clustering lines of the raw blocks into separate bins and detecting boundaries of the bins or segmenting table columns/rows based on alignment for borderless tables.
15 . The dialog tree generation method of claim 10 , wherein the content structure includes titles, headers, footers, headings, and heading trees.
16 . The dialog tree generation method of claim 10 , wherein the content structure is a hierarchical graph data structure in which nodes are organized in a traversable tree.
17 . The dialog tree generation method of claim 10 , wherein the content structure is annotated with action entities defining actions and generic entities defining key phrases.
18 . The dialog tree generation method of claim 10 , wherein the font and typographical properties include at least one of typeface, letterforms, font size, font family, style, color, orientation, length, contrast, position, spacing, or dimension.
19 . A computing system comprising:
a processor, and a memory storing instructions that, when executed by the processor, cause the computing system to implement:
a virtual assistant configured to execute a dialog decision tree to thereby present an interactive chat dialog to a user that proceeds according to the dialog decision tree and includes binary type dialog based on a binary type node branch and an entity based dialog based on an entity based node branching, the virtual assistant further configured to receive a user input, process the user input with the dialog decision tree, and generate an output based on the executed dialog decision tree,
wherein the dialog decision tree is generated by a dialog tree generation software system, configured to:
receive documents;
parse the documents into raw text blocks, raw image blocks, and raw shape blocks;
extract visual design elements from the raw text blocks, the extracted visual design elements including style and geometry properties of the raw blocks;
cluster the raw blocks based on the extracted visual design elements of the raw blocks;
arrange the clustered raw blocks into a hierarchy to generate a document layout model;
process the document layout model to generate a document structure model;
annotate the document structure with entities and synonyms;
generate at least the dialog decision tree based on the annotated document structure, the dialog decision tree comprising a plurality of nodes organized into the hierarchy, and levels of the document structure model correspond to levels of the dialog decision tree, wherein the dialog decision tree includes the binary type node branch from a node to two topics and the entity based node branching from a node to more than two topics; and
output the dialog decision tree to a location accessible by the virtual assistant, wherein the style and geometry properties include at least one of font and typographical properties, paragraph alignment, line lengths, column widths, or graphic shapes.
20 . The computing system of claim 19 , wherein the visual design elements of the document layout model are analyzed to generate the document structure model comprising a hierarchical graph data structure in which nodes are organized in a traversable tree, each node representing a dialog.