TW201426344A - Habit-oriented patent query system proxy - Google Patents

Abstract

The present invention discloses a habit-oriented patent query system proxy, which uses intelligent instruction-operated proxy to fill in the condition values in the field of webpage form and automatically submit to execute the query process. It possesses query technology of automatically field-filling to the original webpage. With the proxy of this query system, in the query of prior art patent information, only key technical terms about the patent to be applied are needed to key in. Further, through the proxy to automatically fill in the constant condition values in each field of the original webpage and finally introduce webpage query sequence of the original search system, it is indeed capable of obtaining the same query results as those obtained directly from the original webpage, which achieves effects of reducing filling process of query fields and enhancing search accuracy and efficiency to improve the query efficiency of a patent query system's user searching for prior arts.

Description

Habit-oriented patent query system agent

The invention is a habit-oriented patent inquiry system agent, which is used to simplify the process of manually inputting the search field by using the agent when using the existing patent information retrieval system, thereby improving the speed and efficiency of patent information retrieval.

Patent is an extremely important part of intellectual property rights. In real life, patent products are ubiquitous. Newspapers related to patents are seen almost every day in newspapers. Before applying for patents, you need to know whether the existing patents of others are related to the patents you have applied for. Repeatedly, professionals engaged in research and development work must also conduct research by collecting patent data of others, so as not to enter the patent mines, and avoid the waste of resources to invest in invalid patents or technology research and development, and the relevant patent application maintenance professionals, Employees of patent offices and employees specializing in patent-related work in the company often use patent enquiry systems in countries and regions on the Internet. They are generally referred to as users of patent enquiry systems. Users of these patent enquiry systems are queried through the patent information retrieval system. In the case of the previous case information, the web page query field is complicated, and the condition values of many fields are often fixed. However, in order to meet the requirements of the web page query conditions, the fixed condition values must still be filled in one field. , causing the searcher to feel the action repeating intolerance and low For efficiency, how to improve the efficiency of the query, the query is the above-mentioned patent system users tried to think about problem solving.

Domestic Patent Application No. 094101774 "The Grammar of Patent Information Retrieval The conversion method discloses a method for automatically converting a search grammar into different search grammars on national patent information websites, the method comprising the steps of: inputting a search string; parsing the input search string; and automatically segmenting the search string After that, a syntax tree is generated; the generated syntax tree is optimized; the optimized individual is subjected to semantic analysis and execution type checking; the grammatical conversion is performed, and the analyzed grammar is translated into different by loading different Schema languages. Search statements; by grammar conversion, automatically generate statements suitable for the online database search grammar of national patent offices, and can also generate SQL search statements and other forms of search statements such as XML language.

Another domestic patent application No. 091137075 "Method for Retrieving Patent Literature Data" discloses that a computer system searches for a patent document related to the keyword, and the computer system further has a keyword category summary table, which makes the key The word is compared with a keyword category summary table to find out which technical categories are likely to belong. Reuse the technical categories that may belong to it, so that some unrelated patent documents can be removed for further retrieval (such as full-text search), or after the search, the user can know which are the more likely related technical categories of patent documents. .

The first pre-existing case mentioned above is that the users of the patent enquiry system can use the friendly interface of the Internet to search for patent information of various countries. However, the search grammar supported by these websites has a considerable degree of difference, so that users Pay attention to different search rules when searching for patent information websites in different countries, which is labor-intensive and reduces efficiency; therefore, the grammar conversion of the patent information retrieval is provided. The method can automatically convert the search grammar into different search grammars on the patent information websites of various countries, so that the user can accurately and quickly search related patent information, and use the same keyword to cross the patent information website system of each country.

The second predecessor is to let the user input at least one keyword into a computer system. The computer system includes a patent document database, so that the computer system finds the patent document related to the keyword, and the computer system has a key. The word category summary table, the technical category of the keyword category summary table record, and the plural feature words corresponding to each technology category, the method needs to first establish a keyword category summary table to make the method work.

In view of the above, a custom-oriented patent query system agent disclosed in the present invention uses a smart instruction-operated agent to fill in the condition value of a web form field and automatically submits it, executes a query program, and reduces the use of patents. The time required to query the system; the query technology with the field automatically filling in the original webpage, through the query system agent, only need to type the key terms to be patented when querying the patent information of the previous case, and then automatically through the agent After filling in the fixed condition value of each field of the original webpage, and finally importing the webpage query thread of the original retrieval system, the same query result can be obtained directly from the original webpage query, thereby reducing the filling process of the query field and improving the retrieval accuracy. Degree and efficiency, which enhances the efficiency of patent query system users in retrieving previous skills; this agent uses and links to existing patent data retrieval systems, as existing patent data retrieval systems have columns for various search conditions. Bit for entering the search key in the relevant field to enter the database for retrieval; Build patent search system agent comprising The following steps: (1) Enter the patent data retrieval system webpage, and use the right mouse button to view the source code, find the relevant fields and variable data, perform web page analysis and field acquisition and setting; (2) use TWebBrowser in Delphi The category constructs the WebBrowser object, which uses the object declared as THTMLDocument to obtain the original code on the webpage, and stores the data in the category of TMemo, which is convenient for observing and obtaining the name of the field and the setting method of the field data; (3) by means of setting Integration completes web page analysis and field acquisition and setting, and completes database configuration to complete programming; (4) intelligent instruction operation design; and (5) agent can be input by completing program integration and testing The word begins to perform the search.

The main purpose of the present invention is to provide a custom-oriented patent inquiry system agent, which can use the existing network patent information inquiry system to input the keyword words to be queried by using the agent of the present invention, and can automatically fill in the original through the agent. The fixed condition value of each field of the webpage is imported into the webpage query thread of the original retrieval system, and the same query result as that of the original webpage query can be obtained, and the time required for inputting keywords and conditions by column is greatly reduced.

The second object of the present invention is to provide a habit-oriented patent query system agent, so that the user of the patent inquiry system can use the agent to simplify the process of inputting keyword words, improve the retrieval efficiency, and achieve the purpose of the two previous cases. Different, can be used together to provide higher efficiency.

A further object of the present invention is to provide a habit-oriented patent query system agent, because according to the establishing step, the commonly used programming language can be used. The agent of the present invention is established so that the user does not increase the cost by using the agent of the present invention, and the relative use of the agent of the present invention can reduce the staff's staff and improve the work efficiency, thus verifying the practicability of the present invention.

First, please refer to the first figure. The first figure is a flowchart of the procedure for constructing the agent of the patent inquiring system of the present invention. The agent of the present invention uses the existing patent data retrieval system and links with the existing patent data. The retrieval system has fields for various search conditions for inputting search keywords in relevant fields to enter the database for retrieval; therefore, the establishment of the patent inquiry system agent includes the following steps: (1) entering the patent data retrieval system webpage, and using Click the right mouse button to view the source code, find the relevant fields and variables, and perform web page analysis and field acquisition and setting. (2) Use the web source code access program to obtain the source code on the web page and store the data. In the category of the field processing program, it is convenient to observe and obtain the field name and field data setting method, and complete the database file creation; (3) complete the web page analysis and field acquisition and setting and complete the database construction through integration File, and then complete the programming; (4) intelligent instruction operation design; and (5) agent that integrates and tests the program. Enter keywords begin word retrieval.

The description of the predetermined work items for further explaining the establishment of the agent of the present invention is as follows:

1, in order to meet the automatic filling of web pages, you can use Microsoft to provide The webBrowser function library is used to perform the field acquisition of the webpage and the automatic filling operation of the data; the scheduled workflow of this item is as shown in the following process 1: document.getElementById("ID in the form").value=''fill in value" Or use the property of Name to handle it.

document.getElementByName("name in the form").value=''fill in value'" Get the submit button in the web page after completion and execute the click to submit the document document.getElementByName("submit button name").click

2. When performing the search, all the conditional field values of the original webpage query conditions must be included. Therefore, a database system is built to record the name of the field in the webpage form and its data content, so that the program can automatically obtain the content of the field and fill in the content. Go to the web form, as shown in the following process 2.

3. In order to avoid the change of the content name of the webpage form, the program is filled in with errors and maintenance troubles. Therefore, the database operation interface needs to be designed, and the input screen is input. Consistent with the content and location of the web form, the need for the operator to compile the web page field and preset values.

4. In order to automate the query operation, and to be able to input all the query form fields in time, it is necessary to design a set of intelligent colloquial instruction operation mode to provide users with relevant conditions (keyword words) according to their needs. . Firstly, the common phrase and instruction code and name are established in the database. The command code is established according to various word classes, such as the verb command starts with 1, the noun starts with 2, and so on. And the grammar rules are established in the database system, mainly for analyzing the input sentences of the user, removing the keywords to obtain the keyword words, and then completing the change of the field value of the query webpage form or performing the query action according to the word meanings. Finally, the result is presented in the screen of the query system agent in text or webpage. The above automated query flow chart is shown in the process 3.

In order to make it easier to understand the execution situation of the query system agent, four query sample pictures are illustrated in the process 4, and the actual query command is in the box on the right side of the input text area.

The first example shows that the agent will have the function of distinguishing between keyword words and keywords. The command is "Querying the computer and network period from 2007 to 2008", in which "query" and "period" will be Interpreted as a query command, "Computer &Network" is a query keyword, and "2007~2008" will be parsed into the year field in the announcement/publication or application date field. Depending on the default value of the Query Agent. In addition, the "and" in the "computer and network" keyword will be translated into the logical operation of AND, and the "yes" will be regarded as the word.

The second query example simply uses the "computer and network" query command. What is the period of the specified query? Therefore, the agent will interpret the keyword words "computer", "AND" and "network". The other query condition fields will be submitted to the original web page for query according to the preset value of the agent.

In addition, the third and fourth query examples show other possible query commands to better understand the capabilities of the query system agent. Basically, the meanings of the first and second examples are not exceeded, so the relevant explanation is omitted here.

In order to further explain, the agent's construction is subdivided into the following five steps, and the goal achievement is described as follows:

(1) Complete web page analysis and field acquisition and setting

Using the well-known patent information webpage established by the Intellectual Property Office of China, first enter the ROC patent system webpage and use the right mouse button to view the source code, as shown in Flow 5.

It can be known from the process 5 that the query buttons of the public/announcement number and the string query are information such as the _IMG_search and _IMG_search 2 and the name of the pull-down menu _1_2_r_3. After detecting the relevant fields and variable data, use the TWebBrowser category in Delphi to construct the WebBrowser object, which uses the object declared as THTMLDocument to get the original code on the webpage, and store the data in the category of TMemo, which is convenient for observation and acquisition. The name of the field and the setting method of the field data. For details, please refer to the process 6.

(2) Complete database file filing

At this point, the web page source code can be taken out, and the code can be written to obtain the data and written into the database field.

After webpage analysis, it is found that there are SELECT, CheckBox, etc. in the content of the webpage, plus the instruction list and the operation status table. Therefore, the database is divided into four data tables, which are:

After confirming the structure of the database and completing the construction, the original webpage of the previously analyzed webpage is stored in the object of Memo1, and the objects of Memo1 can be taken out in a row and a column, and the search object is confirmed to exist. If it exists, the name is found. The location of the Value, using the string to take out the variable name and Value of the Name, write the database to complete the web page analysis and database establishment.

(3) Completing the programming and integrating the functions of (1) and (2) above

The source code for the checkbox and ComBoBox is as follows: <select name=1_2_r_3><option value=’ID’>Publication/Announcement Day

<option value=’AD’>Application date

<option value=’DR’>Priority application date

<option value=’DD’>New technology report completion time

<td><input type=checkbox name=_0_11_s_IDC checked>public/announcement day

Program gets field variables and set values

For i:=0 to memo1.count-1 do

wordPos1:=pos(‘public/announcement Day', memo1.Lines.strings[i]); if wordPos1<>0 then// is the search object

Begin

namePos1:=pos(’name=’,memo1.Lines.strings[i]);//Get the location of the field name

ValuePos1:=pos(’value=’,memo1.Lines.strings[i]);//Get the field setting data

Name:=Copy(memo1.Lines.strings[i], namePos1+5,7);//Get 1_2_r_3

Value:=Copy(memo1.Lines.strings[i],ValuePos1+6,2);End;Adodataset2.Insert;Adodataset2.FieldByName('Name').AsString:=Cpstr[cpidx];Adodataset2.FieldByName('Value ').AsString:=S;Adodataset2.Post;//write data table

After analysis, it is known that the checklist data field is stored in the CheckBox field name, and the field data is as shown in Process 7:

The drop-down data table is for storing the name of the selected field. The field data is as shown in process 8:

Another job is to record the default value of the website, and the program can complete the preset value on the webpage to avoid the repetitive mechanical query action of the searcher and increase the search. The efficiency of the previous case; therefore, in the programming, the same screen as the web page is taken, and the searcher completes the set value of the query condition and automatically writes the status setting data table.

Before the searcher uses the program to query the webpage, the program first inputs the set query conditions into the webpage to complete the setting of the query condition and executes the query instruction.

Set the CheckBox method: WebBrowser1.OleObject.Document.GetElementByID (‘field variable').checked:=0;//Unchecked

WebBrowser1.OleObject.Document.GetElementByID (‘field variable').checked:=1;//check

Set the mode of ComBoBox: Edits:=WebBrowser1.Document.GetElementByID('field variable');Edits.Value:='set value'; Finally, the status setting data table is set in the storage operator to set the query condition variable, field data As shown in the process 9, the process 10 depicts the process of analyzing the variation of the webpage field. The main task is to obtain the name of the field first, and then obtain the position of the variable, and then use the location to obtain the real name of the webpage field and the set variable. Finally, it is written into the database; then it is judged whether there are other field names to be searched, and if so, the above procedure is repeated.

(4) Completing the intelligent instruction operation design

After completing the above-mentioned related materials, the design operation screen is started. The operation screen is deliberately similar to the query page of the Republic of China Patent Information Retrieval System, which reduces the strangeness of the searcher and shortens the learning time. The completed operation screen is as shown in Process 11(a). It is shown that although the field options are similar to the query page of the original Republic of China Patent Information Retrieval System in Process 11(b), it is worth emphasizing that these similar fields will be used by the agent system after the first use of the check state. If you remember, if there is no change in the subsequent query, you do not need to check it again.

(5) Completing the program integration test and writing the report

In order to achieve the query work of the wisdom instruction, the keyword word is set in the data table in advance, and the program will first query the database for the relevant instruction after the commander under the command, and if the relevant instruction is found, the string processing is performed according to the instruction characteristic, and the retrieval is performed. The query keyword of the person is then set to the necessary field in the inquiry page of the Republic of China Patent Information Retrieval System for inquiry.

The instruction data table stores the necessary operation instructions, and the field data is as shown in the process 12; the process 13 is a string instruction query flow chart, and after the query agent obtains the input string, analyzes and determines whether it is a command instruction. If yes, the instruction is stored in the array memory, and then the weight of the string instruction is arranged. First, the keyword is first taken out and placed on the webpage, and finally the query operation of the webpage is performed.

To further illustrate the operation of this Query Broker, the following input string The first example of "Querying the Embedded System" is the first example. Please note that the pre-query patent type shown in Process 14(a) is checked as the invention and the new style, and the new type is not checked. However, after the execution of the agent, the keyword "new type" is first parsed, so the query operation screen automatically changes the option to be new; in addition, the "embedded system" is also parsed and placed in the relevant field, such as the process 14 ( b) shown. Flowchart 14(c) shows the results of the query that brought the query conditions to the original Republic of China patent information retrieval system.

Next, the second example will use "Certificate No. M366253 Invention" as the input query string, and Flow 15(a) shows the operation screen after the input string "Certificate No. M366253 Invention" is parsed, and Flow 15(b) is The query condition is brought to the query result of the original Republic of China patent information retrieval system, and the process 15(c) shows that the query screen automatically displays the content of the patent.

The third example will show the intelligent profiling ability of the query agent for the date query condition. The query "Query M366253 period 20080101 20091231" is an example of the query string. The process 16(a) shows the operation screen after the input string is parsed, and the process 16(b) is the query result of bringing the query condition into the original Republic of China patent information retrieval system.

Through the above-mentioned actual verification operation, it is confirmed that in addition to obtaining the same query result directly from the original webpage query, and the filling process of reducing the query field is achieved, the retrieval accuracy and efficiency are improved, and the application of the patent query agent can be used. The user of the patent inquiry system re-inspires the interest in the query transaction, solves the problem that the user of the patent inquiry system feels that the action is repeated and the work efficiency is low, and at the same time, how to establish a query agent and how to import the webpage program of the original query website. Writing skills.

The above is only a preferred embodiment of the present invention, and it should be noted that those skilled in the art can make several improvements and refinements without departing from the principles of the present invention. Retouching should also be regarded as this The scope of protection of the invention.

In addition, the present invention is used because the agent of the present invention can be established using the commonly used programming language in accordance with the establishing step, so that the user does not increase the cost by using the agent of the present invention, and the relative use of the agent of the present invention can be reduced. When people are employees, the work efficiency is improved, so that the whole industry is practical and cost-effective, and its structure has not been seen in books or public use. It is in line with the requirements for invention patent applications, and please ask the Bureau to grant patents as soon as possible. For the pray.

It is to be understood that the above is a specific embodiment of the present invention and the technical principles applied thereto, and the functional effects produced by the concept of the present invention are still beyond the spirit of the specification and drawings. In the meantime, it should be combined with Chen Ming within the scope of the present invention.

The first figure is a flow chart of the steps of the invention.

Claims (3)

A habit-oriented patent inquiry system agent is linked with an existing patent data retrieval system, and the patent data retrieval system has various search condition fields for inputting search keywords in relevant fields to enter the database for retrieval and construction. The patent enquiry system agent includes the following steps: by entering the patent data retrieval system webpage and using the right mouse button to view the source code, and detecting relevant fields and variable data, performing webpage analysis and field obtaining and setting; using the webpage The source code access program obtains the original code on the webpage, and stores the data in the category of the field processing program, so as to conveniently observe and obtain the field name and field data setting manner, and complete the database file archiving; Completion of web page analysis and field acquisition and setting and completion of database archiving to complete programming; intelligent instruction operation design; and by completing the program integration and testing of the agent, you can enter keyword words to start the search. For example, the patent-oriented patent query system agent described in claim 1 of the patent application, wherein the webpage source code access program used may be an object of the TWebBrowserx category in Delphi. For example, the patent-oriented patent query system agent described in claim 1 of the patent scope, wherein the field processing program used may be an object of the THTMLDocument category.