EP2438507A1 - Procédé, système et appareil de recherche ciblée de documents multi-sectionnels dans un ensemble de documents électroniques - Google Patents

Procédé, système et appareil de recherche ciblée de documents multi-sectionnels dans un ensemble de documents électroniques

Info

Publication number
EP2438507A1
EP2438507A1 EP09844453A EP09844453A EP2438507A1 EP 2438507 A1 EP2438507 A1 EP 2438507A1 EP 09844453 A EP09844453 A EP 09844453A EP 09844453 A EP09844453 A EP 09844453A EP 2438507 A1 EP2438507 A1 EP 2438507A1
Authority
EP
European Patent Office
Prior art keywords
document
collection
vector
intellectual property
static
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
EP09844453A
Other languages
German (de)
English (en)
Other versions
EP2438507A4 (fr
Inventor
Jason David Resnick
Randy W. Lacasse
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
CPA SOFTWARE Ltd
Original Assignee
CPA SOFTWARE Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by CPA SOFTWARE Ltd filed Critical CPA SOFTWARE Ltd
Publication of EP2438507A1 publication Critical patent/EP2438507A1/fr
Publication of EP2438507A4 publication Critical patent/EP2438507A4/fr
Withdrawn legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/93—Document management systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G06F16/3347—Query execution using vector based model

Definitions

  • This invention relates to an electronic document collection, and searching the collection in response to receipt of a query. More specifically, the invention relates to categorizing multiple sections of each document, and efficiently processing the query responsive to the categorized sections of the documents in the collection.
  • An invalidity search may be commissioned to determine if the issued claims of a patent are valid, etc.
  • Prior electronic search tools do not support the different classes of searches. Rather, the burden is on the person doing the search, also known as the searcher, to limit the sections of a patent document to be reviewed in the search based upon the scope of the search. As the quantity of patents and published patent applications in the database grow, the burden on the searches increase as more patents and published patent applications need to be reviewed for each search.
  • This invention comprises a method, system, and article for efficiently and effectively searching a collection of intellectual property documents, such as patent documents.
  • a computer method for searching an electronic document collection.
  • a collection of intellectual property documents is compiled, with each of the intellectual property documents in the collection being comprised of multiple sections.
  • at least one document vector is derived for each patent document in the collection.
  • the derivation of the document vector includes creation of at least one static document vector for each document in the collection.
  • a dynamic document vector is created based upon the string submitted with the query input.
  • submission of the query input to the collection results in a comparison of the dynamic document vector associated with the query input with each static document vector in the collection.
  • a compilation of relevant patent documents are returned based upon a comparison of the dynamic document vector with the static document vectors of the collection.
  • a computer system in communication with storage media, and an electronic document collection maintained on the storage media.
  • the electronic document collection is a compilation of patent or other intellectual property documents. Based upon characteristics of patent documents, each of the patent documents in the collection has multiple sections.
  • At indexing time at least one document vector is derived for each patent document in the collection.
  • the creation of the document vector includes creation of at least one static document vector for each patent document in the document collection.
  • a dynamic document vector is created from string data received from a query input. Following the creation of the dynamic document vector, the query input is submitted to the electronic patent document collection.
  • a query manager in communication with the input manager compares the dynamic document vector to each static document vector in the collection in response to submission of the query input to the patent document collection. Following the submission by the query manager, a compilation of relevant patent documents is returned with the compilation based upon the comparison of the dynamic with the static document vectors.
  • an article is provided with a computer-readable carrier including computer program instructions configured to search an electronic document collection on computer memory.
  • the computer-readable carrier includes computer program instructions to perform over the document collection. Instructions are provided to compile a collection of patent documents. Each of the patent documents in the collection is divided into multiple sections. At the time of indexing the collection, instructions are provided to derive at least one document vector for each patent document in the collection. This includes creation of at least one static document vector for each patent document in the document collection. At the time of submission of a query to the collection, instructions are provided to create a dynamic document vector based on string data from a query input.
  • the query is submitted to the electronic document collection for comparison of the dynamic document vector with each static document vector in the collection.
  • Results of the query submission include a compilation of relevant patent documents returned based upon comparison of the dynamic with the static document vectors in the collection.
  • FIG. 1 is a flow chart illustrating searching an electronic document collection, and more specifically a collection pertaining to patents and patent publications;
  • FIG. 2 is a flow chart illustrating a general process for submission of a query to the patent document collection
  • FIG. 3 is a flow chart illustrating a process for employing stop words to further parse static document vectors in a patent document collection
  • FIG. 4 is a flow chart illustrating a process for creating multiple document vectors for each patent document in the collection
  • FIG. 5 is a flow chart illustrating a process for submission of a query to the document collection with multiple document vectors therein, according to the preferred embodiment of this invention, and is suggested for printing on the first page of the issued patent;
  • FIG. 6 is a block diagram illustrating a set of tools employed to process a query submitted to the electronic document collection.
  • FIG 7. is a block diagram of a graphical user interface for user input designations to search the electronic document collection.
  • a manager may be implemented in programmable hardware devices such as field programmable gate arrays, programmable array logic, programmable logic devices, or the like.
  • the manager may also be implemented in software for execution by various types of processors.
  • An identified manager of executable code may, for instance, comprise one or more physical or logical blocks of computer instructions which may, for instance, be organized as an object, procedure, function, or other construct. Nevertheless, the executables of an identified manager need not be physically located together, but may comprise disparate instructions stored in different locations which, when joined logically together, comprise the manager and achieve the stated purpose of the manager.
  • a manager of executable code could be a single instruction, or many instructions, and may even be distributed over several different code segments, among different applications, and across several memory devices.
  • operational data may be identified and illustrated herein within the manager, and may be embodied in any suitable form and organized within any suitable type of data structure. The operational data may be collected as a single data set, or may be distributed over different locations including over different storage devices, and may exist, at least partially, as electronic signals on a system or network.
  • Static and dynamic document vectors are employed with an intellectual property document.
  • the discussion will be particular to a patent document.
  • the application of the document vectors may be applied to any intellectual property document.
  • a document vector is a set of (keyword, weight) pairs, where the keyword is a word or phrase associated with an underlying document, and the weight is a numerical measure of how important the keyword is for the documents.
  • document vectors are a type of document signature that represents the document content in a manner that facilitates comparison between documents. It is the numerical representation of the unstructured textual content of the document.
  • the static document vectors are associated with patents and published patent applications as these documents are not subject to frequent changes.
  • the dynamic document vector is associated with a query string data, hereinafter strings, submitted to the patent document collection.
  • the static document vectors may be parsed to exclude strings that are specific to patents and have minimal value in conducting a search.
  • the excluded strings are referred to as stop words.
  • the stop words employed herein is specific to the patent community.
  • each patent document has defined sections therein, with each section identifying different portions of a patent document. When conducting a patent search, there are different values placed on the different sections of the patent document. As such, depending upon the scope of the patent search, the search may be limited to specific sections of the patent documents.
  • document vectors are employed in a patent document collection to efficiently and effectively create a result set with data pertinent to the query submitted to the collection, wherein the result set is one or more documents in the patent document collection whose static document vector(s) are calculated to be within a set mathematical range of the dynamic document vector associated with the submitted query string data.
  • Fig. 1 is a flow chart (100) illustrating a general view of searching an electronic document collection, and more specifically a collection pertaining to patents and patent publications.
  • a collection of patent documents is compiled (102). It is recognized in the art that patents and patent publications are comprised of multiple sections.
  • the collection is indexed (104).
  • the process of indexing the compilation includes converting a collection of data into a database suitable for search and retrieval. More specifically, indexing the document collection includes deriving a document vector for each patent document in the collection (106).
  • a document vector comprises a weighted list of words and phrases.
  • terms to be selected into the document vector include, but are not limited, noun phrases, words in title case but not at the beginning of a sentence, and words which occur frequently in the document.
  • Weights are computed for the terms placed into the vector.
  • the following methods for computing the weights may include, but are not limited to, the frequency of the word in the document normalized to a number from one to zero, where one is assigned to the word which occurs most frequently in the document, boosting words or word-pairs in selected fields of the document, assigning a higher weight to noun phrases, elevating title case words in the body of the document, and assigning a higher weight to longer strings over shorter strings.
  • the document vector is computed through employment of an integrator.
  • the integrator can select which fields to include in the vector and how much to boost the words and phrases which they contain, select how much each of the factors contributes to the final term weight, add entity types into the vectors, such as elevating the significance of a corporate entity found in the document, and increasing a stop word list to remove common phrases found in the database.
  • Document vectors created for each patent document in the collection are termed "static document vectors.”
  • the document collection is updated. More specifically, a time interval is established for updating any changes to the documents in the collection, and the associated document vectors (108). Examples of the time interval include, but are not limited to, monthly, semi-annually, annually, etc. Thereafter, it is determined if the established time interval has expired (HO). A positive response to the determination at step ( 110) is followed by a return to step (102).
  • a negative response to the determination at step (110) is following by waiting a set time period to update the patent document vector to incorporate any changes to the patent documents into the document vectors (112), followed by a return to step (110).
  • the patent collection is not limited to granted patents, and includes published patent applications. Accordingly, based upon the inherent nature of patents, a patent document collection should be updated on a periodic basis to address any changes to any of the patents in the collection.
  • Fig. 2 is a flow chart (200) illustrating a general process for submission of a query to the patent document collection.
  • an input query is received (202).
  • the input query is comprised of a string.
  • a document vector is created for the query input (204). Since the document vector for the query is created at the time of submission, it is hereinafter referred to as a dynamic document vector.
  • the dynamic document vector is created based on the text input for the query. More specifically, the dynamic document vector consists of the most relevant terms from the query input text.
  • the following strings are extracted from the input query: noun phrases, words which are in title case, i.e. first letter capitalized but not at the beginning of a sentence, words which occur frequently in the document, pairs of words which occur frequently in the document.
  • designated stop words are removed and not included in the dynamic document vector.
  • weights are assigned to these terms.
  • the frequency of each term or phrase in the document is normalized to a number from 1 to 0, where 1 is assigned to the word which occurs most frequently in the document.
  • words or word-pairs in special fields are boosted, noun phrases are assigned a higher weight, title case words in the body of the document are boosted, longer strings are assigned a higher weight over shorter strings, etc.
  • Computing the document vector is highly configurable.
  • a user can assign a weight to search terms. Accordingly, there are various tools that may be invoked to create an appropriate dynamic document vector based upon the query input.
  • the query in the form of the dynamic document vector is submitted to the document collection (206), where the dynamic document vector is compared to the static document vectors in the patent document collection (208). It is then determined whether any of the static document vectors in the collection are within a defined mathematical range of the dynamic document vector (210).
  • a positive response to the determination at step (210) is followed by placing all of the underlying patent documents in the collection with one or more static document vectors that fall within the defined mathematical range in a result set (212). Either following step (212) or in response to a negative response to the determination at step (210), it is determined if the user would like to submit a new query to the document collection (214). In one embodiment, the new query may narrow the scope of the previously submitted query.
  • the new query may enlarge the scope of the previously submitted query. Regardless of the scope of the new query, a positive response to the determination at step (214) is followed by a return to step (204). Similarly, a negative response to the determination at step (214) marks an end to the query submission process to the document collection. Accordingly, submission of a query to the document collection includes conversion of a submitted string to a dynamic document vector, and comparison of the document vector with the static vectors of the document collection.
  • Patent documents come in the form of issued patent grants and published patent applications. The difference between the two categories of documents identifies their enforceable value. More specifically, a patent grant is an actual property right that can be enforced in a court of law, whereas a published patent application is a pending application that is a pending patent right.
  • Each patent document that is written contains words and phrases that are customary for placement in the application. However, such words and phrases have minimal value in searching, as these words and phrases appears in most patent documents and are not unique to the invention therein. Examples of such words and phrases include, but are not limited to "embodiment”, “exemplary”, “prior art”, etc.
  • each country may have different words that are commonplace in patent applications.
  • the word "characterized” is a common word with little patentable or search value.
  • Such words are referred to herein as stop words.
  • the purpose of identifying stop words specific to a country, language, and or culture, is to minimize the size of the document vectors to be search.
  • Each document vector in the patent document collection may be parsed to remove identified stop words from the collection.
  • Fig. 3 is a flow chart (300) illustrating a process for employing stop words to further parse static document vectors in a patent document collection.
  • the stop words may be limited to a specific country (302), a specific language (304), and/or a specific culture (306).
  • a positive response to any individual selection or combination of selections at steps (302), (304), and/or (306) is followed by creation of a compilation of stop words for parsing the static document vectors in the patent document collection (308).
  • a collection of patent documents is compiled (310). In one embodiment, the collection of patent documents may be limited to the selected country, language, and/or specific culture.
  • the collection is indexed (312) and the stop words are parsed from the collection (314).
  • the process of indexing and removing stop words from the compilation includes converting a collection of data into a database suitable for search and retrieval.
  • one or more sections of the documents in the collection are selected to be included in the document vectors to be created for the collection (316).
  • a document vector is created for each patent document in the collection (318). More specifically, following indexing of the document collection, a document vector is derived for the selected sections of each patent document in the collection with omission of identified stop words from the derived document vectors. Such document vectors are referred to herein as static document vectors.
  • a time interval (320) is established for updating any changes to the documents in the collection, and the associated document vectors. Examples of the time interval include, but are not limited to monthly, semi-annually, annually, etc. Thereafter, it is determined if the established time interval has expired (322). A negative response to the determination at step (322) is followed by waiting a set time period (324) to update the patent document vectors to incorporate any changes to the patent documents into the document vectors, followed by a return to step (320).
  • a positive response to the determination at step (322) is followed by a determination as to whether there are any new stop words to be applied to the document collection (326).
  • a negative response to the determination at step (326) is followed by a return to step (310), and a positive response to the determination at step (326) is followed by adding the new stop word(s) and/or phrase(s) to the compilation of non-relevant patent terms (328).
  • step (328) the process of creating and/or updating static document vectors for a patent document collection returns to step (310). Accordingly, the static document vectors may be parsed for a selection of identified stop words to enable submission of a query to focus on relevant strings in the static document collection.
  • each section of the patent document is required for a submission of a completed patent application, and each section of a patent has a purpose.
  • the details of each section of a patent application are not going to be discussed in detail herein. However, the different sections will be identified.
  • each patent application includes a title, a priority filing date, an abstract, a background description, a summary, a brief description of the drawing figures (if any), a detailed description of the invention, and claims.
  • search categories that are employed in the patent arena depending upon the purpose of the search. For example, an infringement and/or product clearance search is concerned with the words in the claims, and therefore should be directed to the claims present in the document collection.
  • a validity and/or invalidity search is concerned with any known prior art, and requires identification of the priority filing date of the patent document.
  • an inventor(s) seeks to determine the novelty of their invention prior to or following submission of a patent application, the inventors or his/her agent or representative may commission a novelty search.
  • Such a search may de-emphasize the claims and focus on the detailed description of the invention. Accordingly, as shown herein, each search places emphasis on different sections of a patent document in the document collection.
  • each patent in the document collection may be parsed for a selection of stop words that have minimal value in a search of the collection.
  • the creation of multiple document vectors, with each vector identifying a specific section, enables a search of the document collection to be refined based upon a defined scope of the search.
  • an infringement search in the document collection may be limited to document vectors pertaining to the claims section of each patent in the document collection.
  • Fig. 4 is a flow chart (400) illustrating a process for creating multiple document vectors for each patent document in the collection.
  • the collection of patent documents is compiled (402) and indexed (404).
  • the variable M ⁇ ota i is assigned to the total number of documents in the patent document collection (406), and the counting variable M is assigned to the integer one (408).
  • the quantity of sections in patent documents M in the collection is identified (410).
  • the variable Niotai is assigned to the total number of sections in patent document M (412), and the counting variable N is assigned to the integer one (414).
  • a document vector is created for each section of each patent document in the collection. More specifically, a document vector is created for each SectionN of PatentDocumentM (416).
  • the counting variable N is incremented (418) to proceed to the next section of the patent document for creation of the next document vector for the next section, if there is another section of the patent document.
  • a determination is conducted as to whether there are any more sections in the patent document for creation of a document vector (420). A negative response to the determination at step (420) is followed by a return to step (416). Conversely, a positive response to the determination at step (420) is followed by an increment of the variable M (422). It is then determined if each document in the collection has been parsed for creation of multiple document vectors (424).
  • a negative response to the determination at step (424) is followed by a return to step (410) for creation of multiple document vectors for the next document in the collection.
  • the static document collection may need to be updated on a periodic basis.
  • the frequency of the update may be frequent or infrequent depending upon the accuracy of the collection.
  • the frequency of updating the static document vectors may be proportional to the issuance rate of patents.
  • a positive response to the determination at step (424) is an indication that the patent document collection has been parsed to create multiple document vectors for each patent document. It is then determined if the time interval for updating the static vectors in the collection has expired (426).
  • a positive response to the determination at step (426) is followed by a return to step (402).
  • each patent document in the document collection may be parsed to create multiple static document vectors with each vector pertaining to one identified section of the patent document.
  • Fig. 5 is a flow chart (500) illustrating a process for submission of a query to the document collection with multiple document vectors therein.
  • a user submitting a query to the collection defines the scope of the search (502).
  • the user may be provided with a graphical user interface as a layer over computer instructions to facilitate selection of the scope of the search.
  • the defined scope of the search is associated with a selection of document vector categories for the document collection (504), and a query string is submitted to the document collection (506).
  • a dynamic document vector is created for the submitted query string (508), and the dynamic document vector is submitted to the document collection to determine relevant documents (510).
  • the query submission is limited to a comparison of the dynamic document vector with select static document vectors of the document collection (512).
  • the selection of static document vectors may be the selection of a group of static document vectors (513). More specifically, a search that is limited to the claims section of a patent document will only search the static document vectors, or the group of like static document vectors, of the claims section of the patents in the patent document collection.
  • the comparison at step (512) is a mathematical comparison of the dynamic document vector with the static document vectors.
  • a result set of the comparison is sorted based upon the mathematical comparison (514). In one embodiment, the sorting is hierarchical based upon the closeness of the static document vector(s) of the document collection to the dynamic document vector. Accordingly, a comparison of the dynamic document vector with the static document vectors of the collection generates a result set.
  • a mathematical value is employed to define the range of closeness of the sorted documents determined to be relevant (516). Following step (516), it is determined if there are any documents in the sorted collection that fall within the defined mathematical range (518). A positive response to the determination at step (518) is followed by placing a list of all of the underlying patents within a static document vector within the defined range of the dynamic document vector in a result set (520). Following step (520) or a negative response to the comparison at step (518), it is determined if the user wants to submit a new query string or further limit the query of the prior query string submission (522). A negative response to the determination step (522) signals an end to the query submission process.
  • a positive response to the determination at step (522) is followed by a subsequent determination as to whether the user would like to change the sections, i.e. static document vectors, of the search to be compared to the query (524), i.e. dynamic document vector.
  • altering the scope of the search may directly change the selection of static document vectors employed in the search.
  • a positive response to the determination at step (524) is followed by a return to step (502) as the new query will change the sections of the patent document to be evaluated in the next query.
  • a negative response to the determination at step (524) is an indication that the new query will further limit the scope of the prior query while maintaining the limitation of the same document vectors in the patent collection as in the prior query.
  • a negative response is following by submission of the further modification of the query and not the document vectors of the patent document collection, and a return to step (506).
  • the scope of the search may be altered in two aspects to modify the result set based upon the comparison of the dynamic document vector of the query with the static document vectors of the patent document collection.
  • FIG. 6 is a block diagram (600) illustrating a set of tools for creating the static and dynamic document vectors and for employing the vectors in association with a query submitted to the document collection.
  • a computer system 602 is provided with a processor unit (604) coupled to memory (606) by a bus structure (608). Although only one processor unit (604) is shown, in one embodiment, more processor units may be provided in an expanded design.
  • the system (602) is shown in communication with storage media (640) configured to house a document collection (642).
  • the electronic document collection includes a compilation of patent documents, including issued patents and published patent applications.
  • the storage media (640) is in communication with the processor unit (604).
  • the system is shown in communication with a visual display (650) for presentation of visual data.
  • Each of the elements shown and described herein support query submission to the document collection (642).
  • a document manager (660) is provided local to the computer system (602) and in communication with memory (606).
  • the document manager (660) is responsible for deriving a document vector for each patent document in the collection (642) at the time of indexing. More specifically, the document manager (660) creates at least one static document vector (644) for each patent document in the collection (642).
  • each patent document is comprised of specific standardized sections, which may also be uniform if issued from the same patent office jurisdiction.
  • the document manager (660) is employed to create multiple static document vectors (644) for each patent document.
  • the document vectors (644) created by the document manager (660) are housed in the storage media (640).
  • An input manager (662) is also provided local to the computer system (602) and in communication with memory (606).
  • the input manager (662) is responsible for creating a dynamic document vector at query time based on string data received from a query input.
  • the input manager (662) is in communication with a query manager (664), also provided local to the computer system (602) and in communication with memory (606).
  • the query manager (664) is responsible for the comparison of the dynamic document vector, created by the input manager (662), with each static document vector (644) in response to submission of a query input to the document collection (642).
  • the comparison yields a compilation of relevant patent documents (646).
  • the compilation is presented on the visual display (650). Similarly, in one embodiment, the compilation may be retained on storage, either volatile or persistent.
  • a compilation of non-relevant string data (648) may be employed to parse non-relevant string data from the static document vectors (644).
  • the compilation of non-relevant string data (648) is retained on storage media (640) and periodically updated by the document manager (660). Either employing or disregarding the non-relevant string data, the document manager (660) may be directed to create multiple static document vectors for each patent document in the document collection (642).
  • a selection manager (666) is provided local to the computer system (602) and in communication with memory (606). More specifically, the selection manager (666) is in communication with the query manager (664) to select a search scope to the document collection. The selected search scope determines a selection of static document vectors to be applied by the query manager (664) to process the query.
  • the input manager (662), query manager (664), document manager (660), and selection manager (666), may reside in memory (606) local to the computer system (602).
  • the invention should not be limited to this embodiment.
  • the input, query, document, and selection managers (660) - (666) may each reside as hardware tools external to local memory (606), or they may be implemented as a combination of hardware and software.
  • the managers (660) - (666) may reside on a remote system in communication with the storage media (640). Accordingly, a manager may be implemented as a software tool or a hardware tool to support submission of one or more queries to an electronic patent document collection to yield a compilation of relevant patent documents.
  • a query may be submitted to the patent document collection with specific instructions pertaining to the static document vectors to be processed in the query execution.
  • Fig. 7 is a block diagram (700) of a graphical user interface (702) that may be employed to support submission of instructions.
  • the interface (702) functions as a veneer over instructions that support the underlying database of an electronic document collection.
  • the first field (710) includes a field (712) for submission of a query to the document collection.
  • the second field (720) includes multiple fields for selection of a search category.
  • the second field (720) may include the following sub-fields for selection of the search category: novelty (722), state-of-the-art (724), infringement (726), product clearance (728), validity / invalidity (730).
  • the search field (720) may support selection of more than one sub-field.
  • the third field (740) includes multiple fields for selection of the maximum quantity of search documents returned in a result compilation. More specifically, the third filed (740) may include the following sub-fields: ten documents (742), fifty document (744), one hundred documents (746), five hundred documents (748), one thousand documents (750), and an entry field (752) to support customized entry of the maximum quantity to be returned.
  • the invention should not be limited to the sub-field amounts shown at (742) - (750).
  • the fourth field (760) of the interface is employed for submission of the query string to the document collection.
  • the fourth field (760) includes a submit button (762) for entry of the query submission and a cancel button (764) to exit the submission.
  • the interface shown herein facilitates communication and submission of a query to the electronic document collection to leverage the employment of one or more static document vectors therein.
  • the invention is implemented in software, which includes but is not limited to firmware, resident software, microcode, etc.
  • the invention can take the form of a computer program product accessible from a computer-usable or computer-readable medium providing program code for use by or in connection with a computer or any instruction execution system.
  • a computer-usable or computer readable medium can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
  • Embodiments within the scope of the present invention also include articles of manufacture comprising program storage means having encoded therein program code.
  • program storage means can be any available media which can be accessed by a general purpose or special purpose computer.
  • program storage means can include RAM, ROM, EEPROM, CD-ROM, or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired program code means and which can be accessed by a general purpose or special purpose computer. Combinations of the above should also be included in the scope of the program storage means.
  • the medium can be an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system (or apparatus or device) or a propagation medium.
  • Examples of a computer-readable medium include a semiconductor or solid state memory, magnetic tape, a removable computer diskette, random access memory (RAM), read-only memory (ROM), a rigid magnetic disk, and an optical disk.
  • Current examples of optical disks include compact disk B read only (CD-ROM), compact disk B read/write (CD-R/W) and DVD.
  • a data processing system suitable for storing and/or executing program code will include at least one processor coupled directly or indirectly to memory elements through a system bus.
  • the memory elements can include local memory employed during actual execution of the program code, bulk storage, and cache memories which provide temporary storage of at least some program code in order to reduce the number of times code must be retrieved from bulk storage during execution.
  • I/O devices can be coupled to the system either directly or through intervening VO controllers.
  • Network adapters may also be coupled to the system to enable the data processing system to become coupled to other data processing systems or remote printers or storage devices through intervening private or public network.
  • the software implementation can take the form of a computer program product accessible from a computer-useable or computer-readable medium providing program code for use by or in connection with a computer or any instruction execution system.
  • Each patent document is known in the art to have a defined outline of sections that are required to meet statutory filing requirements.
  • Multiple document vectors are created for each individual electronic document with the option to remove non-relevant patent strings from the document vectors.
  • one document vector is created for the claims section of document collection, another document vector is created for the title, abstract, and claims sections of the document collection, and a third document vector is created for all of the section of the document collection combined. Parsing of the vectors yields a smaller and more concise document vector, wherein a smaller document vector improves efficiency of query processing as the vector does not require the additional processing of the parsed strings. Not all queries are the same. Different queries are submitted to the collection to achieve different results. Accordingly, the categorization of the static document vectors, together with parsing of non-relevant patent terms enables a query submission to be efficiently and effectively processed to yield a desirable compilation of document results.
  • searching of intellectual property documents is not limited to granted patents and published patent applications. Searching may be expanded to include all forms of intellectual property documents, including but not limited to trademark registrations and applications, copyright registrations and applications, and all forms of patent documents. Regardless of the document category for the query submission, there is a burden of resources for updating static document vectors in the document collection. Based upon the natural course of the progression of science, the document collection is a growing collection of documents, with new documents added to the collection on a weekly basis or at other times.
  • the time interval set for updating the static document vectors may be a constant as intellectual property documents are granted and published at a set frequency.
  • one or more variables may be employed to change the time interval.
  • the time interval variable may change based upon the quantity of documents that are added to the collection in a defined period of time. The goal is to maintain an accurate document collection that may require periodic updating of the static document vectors in the collection to ensure a comprehensive data repository.
  • the electronic document collection has been specifically described pertaining to intellectual property documents.
  • the invention should not be limited to these specific categories of electronic documents.
  • the electronic document collection may include any type of document that has a defined plurality of sections. This would enable the managers to parse the documents into the defined sections, create multiple static document vectors for each of the defined sections, and support defining a query based upon the defined sections of the documents. Accordingly, the scope of protection of this invention is limited only by the following claims and their equivalents.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Business, Economics & Management (AREA)
  • Computational Linguistics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

L'invention concerne un procédé, un système et un article pour effectuer une recherche efficace et réelle dans un ensemble de documents électroniques. Chacun des documents dans l'ensemble est divisé au préalable en sous-sections, et un vecteur de document statique est créé pour une ou une combinaison de chaque sous-section de chaque document. Un vecteur de document dynamique est créé pour une chaîne d'interrogation soumise à l'ensemble de documents. Sur la base des paramètres de l'interrogation, les sous-sections sélectionnées de chaque document sont utilisées dans une comparaison du vecteur de document dynamique avec les vecteurs de document statiques sélectionnés. Une compilation de documents IP est créée sur la base de tous les vecteurs de document statiques sélectionnés associés qui tombent dans une plage du vecteur de document dynamique.
EP09844453A 2009-05-08 2009-05-08 Procédé, système et appareil de recherche ciblée de documents multi-sectionnels dans un ensemble de documents électroniques Withdrawn EP2438507A4 (fr)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/US2009/043371 WO2010128974A1 (fr) 2009-05-08 2009-05-08 Procédé, système et appareil de recherche ciblée de documents multi-sectionnels dans un ensemble de documents électroniques

Publications (2)

Publication Number Publication Date
EP2438507A1 true EP2438507A1 (fr) 2012-04-11
EP2438507A4 EP2438507A4 (fr) 2013-03-20

Family

ID=43050307

Family Applications (1)

Application Number Title Priority Date Filing Date
EP09844453A Withdrawn EP2438507A4 (fr) 2009-05-08 2009-05-08 Procédé, système et appareil de recherche ciblée de documents multi-sectionnels dans un ensemble de documents électroniques

Country Status (8)

Country Link
EP (1) EP2438507A4 (fr)
JP (1) JP5516916B2 (fr)
KR (1) KR20140056402A (fr)
CN (1) CN102804125A (fr)
AU (1) AU2009345829A1 (fr)
CA (1) CA2761542A1 (fr)
NZ (1) NZ596910A (fr)
WO (1) WO2010128974A1 (fr)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5627750B1 (ja) * 2013-09-11 2014-11-19 株式会社Ｕｂｉｃ 文書分析システム及び文書分析方法並びに文書分析プログラム
WO2015145524A1 (fr) * 2014-03-24 2015-10-01 株式会社Ｕｂｉｃ Système d'analyse de document, procédé d'analyse de document et programme d'analyse de document
JP2015056185A (ja) * 2014-09-30 2015-03-23 株式会社Ｕｂｉｃ 文書分析システム及び文書分析方法並びに文書分析プログラム
EP3716099A4 (fr) 2017-11-22 2021-07-14 Kao Corporation Dispositif de classification de documents
CN111078730A (zh) * 2019-12-23 2020-04-28 广东聚智诚科技有限公司 一种基于知识产权新颖性提取建立用户需求库的系统及方法

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6339767B1 (en) * 1997-06-02 2002-01-15 Aurigin Systems, Inc. Using hyperbolic trees to visualize data generated by patent-centric and group-oriented data processing
US6038561A (en) * 1996-10-15 2000-03-14 Manning & Napier Information Services Management and analysis of document information text
US8095581B2 (en) * 1999-02-05 2012-01-10 Gregory A Stobbs Computer-implemented patent portfolio analysis method and apparatus
JP4497337B2 (ja) * 2000-06-29 2010-07-07 株式会社野村総合研究所 概念検索装置およびコンピュータプログラムを記録した記録媒体
US6662178B2 (en) * 2001-03-21 2003-12-09 Knowledge Management Objects, Llc Apparatus for and method of searching and organizing intellectual property information utilizing an IP thesaurus
US9235849B2 (en) * 2003-12-31 2016-01-12 Google Inc. Generating user information for use in targeted advertising
JP2007018186A (ja) * 2005-07-06 2007-01-25 Shigematsu:Kk 権利調査支援システム
JPWO2008004563A1 (ja) * 2006-07-03 2009-12-03 株式会社アイ・ピー・ビー 研究者求人求職マッチングシステム及び共同研究／共同事業マッチングシステム
JPWO2008075744A1 (ja) * 2006-12-20 2010-04-15 株式会社パテント・リザルト 情報処理装置、提携先を選定するための情報を生成する方法、およびプログラム

Non-Patent Citations (8)

* Cited by examiner, † Cited by third party
Title
BAEZA-YATES R ET AL: "Modern Information Retrieval, CLASSIC INFORMATION RETRIEVAL", 1 January 1999 (1999-01-01), MODERN INFORMATION RETRIEVAL, ACM PRESS, NEW YORK, PAGE(S) 24,25, XP002262943, ISBN: 978-0-201-39829-8 * the whole document * *
GONZALO NAVARRO ED - BAEZA-YATES R ET AL: "MODERN INFORMATION RETRIEVAL, Chapter 8: Indexing and Searching", 1 January 1999 (1999-01-01), MODERN INFORMATION RETRIEVAL, ACM PRESS, NEW YORK, PAGE(S) 191 - 228, XP002457291, ISBN: 978-0-201-39829-8 * chapter 8.2 * *
HISAO MASE ET AL: "Proposal of two-stage patent retrieval method considering the claim structure", ACM TRANSACTIONS ON ASIAN LANGUAGE INFORMATION PROCESSING, vol. 4, no. 2, 1 June 2005 (2005-06-01), pages 190-206, XP055052603, ISSN: 1530-0226, DOI: 10.1145/1105696.1105702 *
JAE-HO KIM ET AL: "Patent Document Retrieval and Classification at KAIST", PROCEEDINGS OF NTCIR-5 WORKSHOP, 1 January 2005 (2005-01-01), XP055052773, Tokyo, Japan ISBN: 978-4-86-049033-1 *
RICARDO BAEZA-YATES: "An Extended Model for Full Text Databases", JOURNAL OF BRAZILIAN CS SOCIETY, vol. 3, no. 2, 1 January 1996 (1996-01-01) , pages 57-64, XP055052770, *
See also references of WO2010128974A1 *
TORU TAKAKI ET AL: "Associative document retrieval by query subtopic analysis and its application to invalidity patent search", PROCEEDINGS OF THE THIRTEENTH ACM CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT , CIKM '04, 1 January 2004 (2004-01-01), page 399, XP055032925, New York, New York, USA DOI: 10.1145/1031171.1031251 *
XUE-FENG WANG ET AL: "Chinese Patent Infringement Retrieval Model Based on the Structure Information of Patent's Claims", MODELLING, SIMULATION AND OPTIMIZATION, 2008. WMSO '08. INTERNATIONAL WORKSHOP ON, IEEE, PISCATAWAY, NJ, USA, 27 December 2008 (2008-12-27), pages 241-246, XP031410201, ISBN: 978-0-7695-3484-8 *

Also Published As

Publication number Publication date
JP5516916B2 (ja) 2014-06-11
JP2012526319A (ja) 2012-10-25
KR20140056402A (ko) 2014-05-12
NZ596910A (en) 2014-02-28
EP2438507A4 (fr) 2013-03-20
CA2761542A1 (fr) 2010-11-11
WO2010128974A1 (fr) 2010-11-11
CN102804125A (zh) 2012-11-28
AU2009345829A1 (en) 2012-01-12

Similar Documents

Publication Publication Date Title
US20100287148A1 (en) Method, System, and Apparatus for Targeted Searching of Multi-Sectional Documents within an Electronic Document Collection
EP0970428B1 (fr) Systeme et methode de classification automatique de documents
JP5534266B2 (ja) 電子文書コレクションからクエリ結果を送付する方法、システム及び装置
US20070168946A1 (en) Collaborative software development systems and methods providing automated programming assistance
US8364679B2 (en) Method, system, and apparatus for delivering query results from an electronic document collection
US20180004838A1 (en) System and method for language sensitive contextual searching
US20100287177A1 (en) Method, System, and Apparatus for Searching an Electronic Document Collection
WO2019171190A1 (fr) Système et procédé de recherche sur la base de blocs de texte et opérateurs de recherche associés
EP2438507A1 (fr) Procédé, système et appareil de recherche ciblée de documents multi-sectionnels dans un ensemble de documents électroniques
EP2427830B1 (fr) Procédé, système et appareil de recherche d'un ensemble de documents électroniques
US8862586B2 (en) Document analysis system
US20040186833A1 (en) Requirements -based knowledge discovery for technology management
CN109783650B (zh) 中文网络百科知识去噪方法、系统及知识库
Hwang et al. System for extracting domain topic using link analysis and searching for relevant features
WO2018220688A1 (fr) Générateur de dictionnaire, procédé de génération de dictionnaire, et programme
Hirsch et al. Evolving Lucene search queries for text classification
WO2015125088A1 (fr) Procédé de caractérisation de document
KR100932046B1 (ko) 도서 검색 방법 및 도서 검색 시스템
Zubarev et al. Method for Expert Search Using Topical Similarity of Documents
Grigoriev Method for Expert Search Using Topical Similarity of Documents
Krabben Machine Learning vs. Knowlegde Engineering in Classification of Sentences in Dutch Law
EP1643379B1 (fr) Système de recherche de document
Denecke et al. Topic Classification Using Limited Bibliographic Metadata

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20111208

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO SE SI SK TR

DAX Request for extension of the european patent (deleted)
A4 Supplementary search report drawn up and despatched

Effective date: 20130215

RIC1 Information provided on ipc code assigned before grant

Ipc: G06F 17/30 20060101AFI20130211BHEP

17Q First examination report despatched

Effective date: 20160118

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN

18D Application deemed to be withdrawn

Effective date: 20160729