Classifying input fields and groups of input fields of a webpage
A method, a system and a machine-readable data carrier, which are intended to classify input fields and groups of input fields of a webpage, are described. The method for classifying input fields and groups of input fields of a webpage may include a browser extension that contains program elements Content Script and Background Script and a program recognition module. The method may form a hierarchy of HTML-documents of a webpage. A multicomponent system for classifying input fields and groups of input fields of a webpage may include a processor device and a storage, and a set of instructions that, when executed by the processor, control the processor to form a hierarchy with grouped fields and a report file for HTML-documents of the webpage. The instructions may also identify elements of the HTML-documents on the webpage.
1. A computer-implemented method comprising:
identifying content of a webpage, the webpage comprising input fields, the webpage comprising two or more HTML-encoded documents, the identifying comprising:
executing, on one or more processors and for each HTML-encoded document of the two or more HTML-encoded documents, a separate instance of first computer-executable instructions, wherein each instance comprises:
generating, for each HTML-encoded document of the two or more HTML-encoded documents, a document identifier associated with the HTML-encoded document;
determining, based on the document identifiers, a hierarchy of the HTML-encoded documents;
determining, for each HTML-encoded document, important and visible elements; and
generating, based on the hierarchy and the important and visible elements, a report;
executing, on the one or more processors and based on the report, second computer-executable instructions that generate a tree data structure of the important and visible elements; and
outputting the tree data structure.
2. The computer-implemented method of claim 1 ,
wherein executing the second computer-executable instructions comprise determining, via a machine-learning model and as important and visible elements, clickable elements.
3. The computer-implemented method of claim 1 , wherein executing the second computer-executable instructions comprises:
determining, via a machine-learning model and for the important and visible elements, label classifications.
4. The computer-implemented method of claim 1 , wherein executing the second computer-executable instructions comprises:
determining, via a machine-learning model, a probability that a label is associated with an input field.
5. The computer-implemented method of claim 1 , wherein executing the second computer-executable instructions comprises:
determining, via a machine-learning model, a tree graph with feature vectors.
6. The computer-implemented method of claim 1 , wherein executing the second computer-executable instructions comprises:
generating, via a machine-learning model, a JavaScript Object Notation (JSON) document comprising a tree graph with feature vectors.
7. The computer-implemented method of claim 6 , wherein determining important and visible elements comprises:
determining, for each for each HTML-encoded document, one or more clickable elements.
8. The computer-implemented method of claim 1 , wherein determining important and visible elements comprises:
determining, for each HTML-encoded document, one or more form submission buttons.
9. The computer-implemented method of claim 1 , wherein determining important and visible elements comprises:
determining, for each for each HTML-encoded document, a label associated with an input field of the input fields.
10. The computer-implemented method of claim 9 , wherein determining important and visible elements comprises determining, for each for each HTML-encoded document, one or more form submission buttons, and the computer-implemented method further comprising:
grouping, into one or more groups, the input fields with the one or more form submission buttons.
11. The computer-implemented method of claim 10 , further comprising:
determining one or more labels for the groups.
12. The computer-implemented method of claim 11 , further comprising:
classifying, based on a machine-learning model, the input fields.
13. The computer-implemented method of claim 12 , further comprising:
populating, based on the classifying, the input fields; and
outputting the populated input fields.
14. An apparatus comprising:
one or more processors; and
memory storing instructions that, when executed by the one or more processors, cause the apparatus to:
identify content of a webpage, the webpage comprising input fields, the webpage comprising two or more HTML-encoded documents, wherein the instructions to identify cause the apparatus to:
execute, on one or more processors and for each HTML-encoded document of the two or more HTML-encoded documents, a separate instance of first computer-executable instructions, wherein each instance comprises:
generate, for each HTML-encoded document of the two or more HTML-encoded documents, a document identifier associated with the HTML-encoded document;
determine, based on the document identifiers, a hierarchy of the HTML-encoded documents;
determine, for each HTML-encoded document, important and visible elements; and
generate, based on the hierarchy and the important and visible elements, a report;
execute, on the one or more processors and based on the report, second computer-executable instructions that generate a tree data structure of the important and visible elements; and
output the tree data structure.
15. The apparatus of claim 14 , wherein the execution of the second computer-executable instructions cause the apparatus to:
determine, via a machine-learning model and as important and visible elements, clickable elements.
16. The apparatus of claim 14 , wherein the execution of the second computer-executable instructions cause the apparatus to:
determine, via a machine-learning model and for the important and visible elements, label classifications.
17. The apparatus of claim 14 , wherein the execution of the second computer-executable instructions cause the apparatus to:
determine, via a machine-learning model, a probability that a label is associated with an input field.
18. The apparatus of claim 14 , wherein the execution of the second computer-executable instructions cause the apparatus to:
determine, via a machine-learning model, a tree graph with feature vectors.
19. The apparatus of claim 14 , wherein the execution of the second computer-executable instructions further cause the apparatus to:
determine, via a machine-learning model, a tree graph with feature vectors.
20. The apparatus of claim 14 , wherein the instructions to determine important and visible elements further causes the apparatus to:
determine, for each HTML-encoded document, one or more clickable elements.