# US 9,519,727 B2 · Improving the efficiency of real-time website optimization processes using saliency analysis

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 9,519,727 B2
- Document ID: US9519727B2
- Status: Granted
- Kind: B2
- Application number: 13/986,973
- Filed: 2013-06-20
- Published: 2014-12-25
- Granted: 2016-12-13
- Earliest claimed: 2013-06-20
- Assignees: Hiconversion, Inc.
- Applicants: Zijad F. Aganovic
- Inventors: Zijad F. Aganovic
- Web page: https://patentgrep.com/patent/US9519727B2

## CPC

- G06F 16/957: Browsing optimisation, e.g. caching or content distillation
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

In order to improve the efficiency of website optimization processes a novel pre-processing method at least includes: a) receiving website source code or web browser code from a website to be optimized; b) converting the website source code or web browser code into an electronic image; c) mapping web page elements corresponding to areas in the image; d) performing saliency analysis on the image; e) mapping the results of the saliency analysis to the web page elements; and f) selecting at least one web page for inclusion in a website optimization process based on the saliency analysis. A corresponding novel system is also disclosed.

## Claims

### Claim 1

A method of pre-processing for website optimization comprising:
a) receiving website source code or web browser code from a website to be optimized;
b) converting said website source code or web browser code into an electronic image comprising a bitmap;
c) mapping web page elements corresponding to areas in said image mapping nodes of the logical tree to corresponding areas of said bitmap image;
d) performing saliency analysis on the said image;
e) mapping the results of said saliency analysis to said web page elements;
f) selecting at least one web page for inclusion in a website optimization process based on said saliency analysis;
g) parsing the code of element b);
h) generating a web page logical tree from the code of element b); and
i) mapping nodes of the logical tree to corresponding areas of said bitmap image.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
generating optimization variables to be utilized in the website optimization process.

### Claim 3 (depends on 1)

The method of claim 1, wherein element f) further comprises:
eliminating from consideration in the website optimization process, web pages containing web page element combinations which are contrary to predefined saliency expectations or website optimization variables that are contrary to predefined saliency expectations.

### Claim 4 (depends on 1)

The method of claim 1, further comprising:
receiving website optimization process results;
performing secondary saliency analysis on said website optimization process results; and
supplying the results of said secondary saliency analysis for a new website optimization process.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
allocating a saliency result area to a specific section of a web page image according to a predefined mapping rules and according to the overlap of the saliency result area with respect to the specific section; and
mapping the saliency of the specific section to a corresponding element of the web page or web browser code.

### Claim 6 (depends on 1)

The method of claim 1, wherein the elements are performed off-line.

### Claim 7

A method of pre-processing for website optimization comprising:
a) receiving website source code or web browser code from a website to be optimized;
b) converting said website source code or web browser code into an electronic image;
c) mapping web page elements corresponding to areas in said image;
d) performing saliency analysis on the said image;
e) mapping the results of said saliency analysis to said web page elements;
f) selecting at least one web page for inclusion in a website optimization process based on said saliency analysis;
receiving website optimization process results;
performing secondary saliency analysis on said website optimization process results;
comparing the results of said secondary saliency analysis with a predefined saliency computation model;
in response to comparing the results of said secondary saliency analysis with a predefined saliency computation model, generating discrepancy indicia; and
modifying said predefined saliency computation model in response to said discrepancy indicia.

### Claim 8 (depends on 7)

The method of claim 7, further comprising:
allocating a saliency result area to a specific section of a web page image according to a predefined mapping rules and according to the overlap of the saliency result area with respect to the specific section; and
mapping the saliency of the specific section to a corresponding element of the web page or web browser code.

### Claim 9 (depends on 7)

The method of claim 7, wherein the elements are performed off-line.