# US 10,482,149 B2 · Method and system for classification of web browsing history > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,482,149 B2 - Document ID: US10482149B2 - Status: Granted - Kind: B2 - Application number: 15/464,999 - Filed: 2017-03-21 - Published: 2018-09-13 - Granted: 2019-11-19 - Earliest claimed: 2017-03-09 - Assignees: Wipro Limited - Applicants: Wipro Limited - Inventors: Suraj Gururaj - Web page: https://patentgrep.com/patent/US10482149B2 ## CPC - G06F 16/955: using information identifiers, e.g. uniform resource locators [URL] - G06F 16/128: {Details of file system snapshots on the file-level, e.g. snapshot creation, administration, deletion (error detection or correction of the data by redundancy in operations or in hardware G06F11/14, G06F11/16)} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract The present disclosure relates to method and computing device for classification of web browsing history by classification system. The classification system receives web browsing history from web browser associated with user, where web browsing history comprises details about one or more web pages browsed by user, extracts one or more keywords from each of one or more web pages browsed by user based on trained keyword dataset, determines a plurality of classifications for each of the one or more web pages based on one or more keywords, generates relevancy matrix between one or more keywords of web pages and corresponding plurality of classifications and identifies a classification from plurality of classifications for each of one or more webpages based on relevancy matrix, where snapshot of classification is stored in non-volatile storage unit of web browser. The use of non-volatile storage unit in present disclosure provides no restriction on storage space. ## Claims ### Claim 1 A method for classification of web browsing history, the method comprising: receiving, by a classification computing device, the web browsing history from a web browser associated with a user, wherein the web browsing history comprises details about one or more web pages browsed by the user; extracting, by the classification computing device, one or more keywords from each of the one or more web pages browsed by the user based on a trained keyword dataset; determining, by the classification computing device, a plurality of classifications for each of the one or more web pages based on the one or more keywords, wherein the plurality of classifications associated with each of the one or more webpages are determined by identifying common characteristics between the one or more keywords and parameters of each web browsing path using a trained classification dataset; generating, by the classification computing device, a relevancy matrix between each of the one or more keywords of the one or more web pages and the corresponding plurality of classifications; and identifying, by the classification computing device, a classification from the plurality of classifications for each of the one or more webpages based on the relevancy matrix, wherein a snapshot of the classification is stored in a non-volatile storage unit of the web browser. ### Claim 2 (depends on 1) The method as claimed in claim 1, further comprising identifying a classification for the one or more web pages browsed by the user based on the snapshot of the classification stored in the non-volatile storage unit of the web browser. ### Claim 3 (depends on 1) The method as claimed in claim 1, wherein the details about the one or more web pages comprises metadata of the web page, data regarding header, Uniform Resource Locator (URL), title and time stamp in the one or more web pages browsed by the user. ### Claim 4 (depends on 1) The method as claimed in claim 1, wherein the web browsing history is received from the web browser at pre-defined time intervals. ### Claim 5 (depends on 1) The method as claimed in claim 1, wherein extracting the one or more keywords comprises: providing, by the classification system, the details about the one or more web pages to a trained keyword extraction model, wherein the trained keyword extraction model is trained using a plurality of training web page dataset; and extracting, by the classification system, the one or more keywords for each of the one or more web pages based on analysis of the trained keyword extraction model. ### Claim 6 (depends on 1) The method as claimed in claim 1, wherein the trained classification dataset comprises one or more classification identified previously based on the web browsing history of the user. ### Claim 7 (depends on 1) The method as claimed in claim 1 further comprising adding a user defined keyword to the trained keyword dataset. ### Claim 8 (depends on 1) The method as claimed in claim 1, wherein the relevancy matrix is generated by calculating a cosine similarity score between the one or more keywords of the one or more web pages and the corresponding plurality of classifications. ### Claim 9 (depends on 1) The method as claimed in claim 1, wherein the snapshot of the classification comprises details about time stamp, URL, title, cosine similarity score and relevancy matrix for each of the one or more web pages. ### Claim 10 A classification computing device comprising: a processor; and a memory communicatively coupled to the processor, wherein the memory stores processor instructions, which, on execution, causes the processor to: receive the web browsing history from a web browser associated with a user, wherein the web browsing history comprises details about one or more web pages browsed by the user; extract one or more keywords from each of the one or more web pages browsed by the user based on a trained keyword dataset; determine a plurality of classifications for each of the one or more web pages based on the one or more keywords, wherein the plurality of classifications associated with each of the one or more webpages are determined by identifying common characteristics between the one or more keywords and parameters of each web browsing path using a trained classification dataset; generate a relevancy matrix between each of the one or more keywords of the one or more web pages and the corresponding plurality of classifications; and identify a classification from the plurality of classifications for each of the one or more webpages based on the relevancy matrix, wherein a snapshot of the classification is stored in a non-volatile storage unit of the web browser. ### Claim 11 (depends on 10) The classification computing device as claimed in claim 10, wherein the processor identifies a classification for the one or more web pages browsed by the user based on the snapshot of the classification stored in the non-volatile storage unit of the web browser. ### Claim 12 (depends on 10) The classification computing device as claimed in claim 10, wherein the details about the one or more web pages comprises metadata of the web page, data regarding header, Uniform Resource Locator (URL), title and time stamp in the one or more web pages browsed by the user. ### Claim 13 (depends on 10) The classification computing device as claimed in claim 10, wherein the processor receives the web browsing history from the web browser at pre-defined time intervals. ### Claim 14 (depends on 10) The classification computing device as claimed in claim 10, wherein the processor extracts the one or more keywords by: providing the details about the one or more web pages to a trained keyword extraction model, wherein the trained keyword extraction model is trained using a plurality of training web page dataset; and extracting the one or more keywords for each of the one or more web pages based on analysis of the trained keyword extraction model. ### Claim 15 (depends on 10) The classification computing device as claimed in claim 10, wherein the trained classification dataset comprises one or more classification identified previously based on the web browsing history of the user. ### Claim 16 (depends on 10) The classification computing device as claimed in claim 10, wherein the processor generates the relevancy matrix by calculating a cosine similarity score between the one or more keywords of the one or more web pages and the corresponding plurality of classifications. ### Claim 17 (depends on 10) The classification computing device as claimed in claim 10, wherein the snapshot of the classification comprises details about time stamp, URL, title, cosine similarity score and relevancy matrix for each of the one or more web pages. ### Claim 18 A non-transitory computer readable medium including instruction stored thereon that when processed by at least one processor cause a classification computing device to perform operations comprising: receiving the web browsing history from a web browser associated with a user, wherein the web browsing history comprises details about one or more web pages browsed by the user; extracting one or more keywords from each of the one or more web pages browsed by the user based on a trained keyword dataset; determining a plurality of classifications for each of the one or more web pages based on the one or more keywords, wherein the plurality of classifications associated with each of the one or more webpages are determined by identifying common characteristics between the one or more keywords and parameters of each web browsing path using a trained classification dataset; generating a relevancy matrix between each of the one or more keywords of the one or more web pages and the corresponding plurality of classifications; and identifying a classification from the plurality of classifications for each of the one or more webpages based on the relevancy matrix, wherein a snapshot of the classification is stored in a non-volatile storage unit of the web browser.