WO2007103096A2 - Procédés et systèmes relatifs aux moteurs de recherche pour l'affichage de sujets pertinents - Google Patents

Procédés et systèmes relatifs aux moteurs de recherche pour l'affichage de sujets pertinents Download PDF

Info

Publication number
WO2007103096A2
WO2007103096A2 PCT/US2007/005222 US2007005222W WO2007103096A2 WO 2007103096 A2 WO2007103096 A2 WO 2007103096A2 US 2007005222 W US2007005222 W US 2007005222W WO 2007103096 A2 WO2007103096 A2 WO 2007103096A2
Authority
WO
WIPO (PCT)
Prior art keywords
topics
topic
search
information
words
Prior art date
Application number
PCT/US2007/005222
Other languages
English (en)
Other versions
WO2007103096A3 (fr
Inventor
Paul S. Odom
Dennis Kotlar
Original Assignee
Kang Jo Mgmt. Limited Liability Company
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 Kang Jo Mgmt. Limited Liability Company filed Critical Kang Jo Mgmt. Limited Liability Company
Priority to JP2008557356A priority Critical patent/JP2009528630A/ja
Publication of WO2007103096A2 publication Critical patent/WO2007103096A2/fr
Publication of WO2007103096A3 publication Critical patent/WO2007103096A3/fr

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
    • G06F16/95—Retrieval from the web
    • G06F16/951—Indexing; Web crawling techniques

Definitions

  • the present invention relates to search engines, and more particularly, to search engine methods and systems that provide relevant and timely topics.
  • Illustrative supervised classification technologies include semantic networks and neural networks. While supervised systems generally derive classifications more attuned to what a human would generate, they often require substantial training and tuning by expert operators and, in addition, often rely for their results on data that is more consistent or homogeneous that is often possible to obtain in practice. Hybrid systems attempt to fuse the benefits of manual classification methods with the speed and processing capabilities employed by unsupervised and supervised systems. In known hybrid systems, human operators are used to derive "rules of thumb" which drive the underlying classification engines.
  • the boss would like the individual to send a copy of the email and the references back to him as soon as possible. Also, he would like the individual to check for additional references to see if the conclusions in the memo need to be updated.
  • the boss requires that the project be completed within fifteen minutes.
  • the worker is not disorganized, but as is common, does not have total recall of how the information was gathered or where the email is stored. After thirty minutes, the worker finally finds the email. But, the worker still needs to search for additional information as requested by his boss. The end result is that because no efficient search mechanism existed the worker has missed his boss' deadline.
  • search engines While a search engine may have these topics in its database, these topics would be part of tens of thousands of possible topic results for a query using the term "NFL.”
  • NNL the "Nenver vs. Pittsburgh” and “Charlotte vs. Seattle” topics would likely be a very meaningful topic result.
  • search engines do not directly factor in time relevancy, and these topics would be mixed in with the tens of thousands of other possible topic results. Thus, a user would not likely receive as relevant search results as would be desired.
  • Another shortcoming of current search engines that display topics or search results is that search engines do not display topics associated with every subject matter domain related to a search constraint entered by a user.
  • a search engine may only show search results or topics that are most popular without regard to different subject matter domains that search results may belong to.
  • the data items belonging to the search results may include topics that correspond to subject matter domains that include autos (e.g., there is a car named Jaguar), animals (e.g., there is an animal called Jaguar), software (e.g., there is a software package referred to as Jaguar), resorts (e.g., there are resorts in South America referred to as Jaguar resorts), football (e.g., there is a football team referred to as the Jacksonville Jaguars) and game (e.g., there is a game referred to a Jaguar).
  • autos e.g., there is a car named Jaguar
  • animals e.g., there is an animal called Jaguar
  • software e.g., there is a software package referred to as Jaguar
  • resorts e.g., there are resorts in South America referred to as Jaguar resorts
  • football e.g., there is
  • search engines that provide results based only on popularity of website hits, might only display topics or search results associated with the subject matter domain Auto. Or, at the very least, items associated with Resorts would be on page 27 of the search results. More often than not, a user probably would be looking for data items in the subject matter domain Auto. However, a reasonable proportion of users may also be interested in other domains that may be less popular. For these users, the search results displayed would not be particularly relevant and their specific areas of interest difficult to find. Thus, a user once again may not receive search results relevant to their particular area of interest.
  • search methods and systems that can efficiently generate search results to identify and display topics by considering, at any given time, the relative significance of a topic based on current events and that ensure coverage of all subject matter domains associated with a search constraint.
  • FIG. 1 is a flowchart of a method to identify topics in a corpus of data in accordance with one embodiment of the invention.
  • FIG. 2 is a flowchart of a method to generate a domain specific word list in accordance with one embodiment of the invention.
  • FIG. 3 is a flowchart of a method to identify topics in a corpus of data in accordance with one embodiment of the invention.
  • FIG. 4 is a flowchart of a method to measure actual usage of significant words in a corpus of data in accordance with one embodiment of the invention.
  • FIG. 5 is a flowchart of a topic refinement process in accordance with one embodiment of the invention.
  • FIGS. 6 is a flowchart of a topic identification method in accordance with one embodiment of the invention.
  • FIG. 7 is a flowchart of one method in accordance with the invention to identify those topics for display during a user query operation.
  • FIG. 8 is a diagram that shows enterprise information sources.
  • FIG. 9 is a flowchart of a method ; for displaying topics, according to an embodiment of the invention.
  • FIG. 10 provides a screen shot of a search engine web site, according to an embodiment of the invention.
  • FIG. 11 is a flowchart of a method for displaying topics, according to an embodiment of the invention.
  • FIG. 12 is a flowchart of a method for displaying topics, according to an embodiment of the invention.
  • FIG. 13 is a flowchart of a method to rank topics into one of four general rankings, according to an embodiment of the invention.
  • FIG. 14 is a diagram that illustrates topic clustering, according to an embodiment of the invention.
  • FIG. 15 is a block diagram of a system, according to an embodiment of the invention.
  • a collection of topics is determined for a first corpus of data, wherein the topics are domain specific, based on a statistical analysis of the first data, corpus and substantially automatically generated.
  • the topics may be associated with each "segment" of a second corpus of data, wherein a segment is a user-defined quantum of information.
  • Example segments include, but are not limited to, sentences, paragraphs, headings (e.g., chapter headings, titles of manuscripts, titles of brochures and the like), chapters and complete documents.
  • Data comprising the data corpus may be unstructured (e.g., text) or structured (e.g., spreadsheets and database tables).
  • topics may be used during user query operations to return a result set based on a user's query input.
  • one method in accordance with the invention uses domain specific word list 100 as a starting point from which to analyze data 105 (block 110) to generate domain specific topic list 115.
  • topic list 115 entries may be associated with each segment of data 105 (block 120) and stored in database 125 where it may be queried by user 135 through user interface 130.
  • Word list 100 may comprise a list of words or word combinations that are meaningful to the domain from which data 105 is drawn. For example, if data 105 represents medical documents then word list 100 may be those words that are meaningful to the medical field or those sub fields within the field of medicine relevant to data 105.
  • Data 105 may be substantially any form of data, structured or unstructured.
  • data 105 comprises unstructured text files such as medical abstracts and/or articles,
  • data 105 comprises books, newspapers, magazine content or a combination of these sources.
  • data 105 comprises structured data such as design documents and spreadsheets describing an oil refinery process.
  • data 105 comprises content tagged image data, video data and/or audio data.
  • data 105 comprises a combination of structured and unstructured data.
  • Data 105 may also include data gathered from across a network, such as the Internet. Acts in accordance with block 110 use word list 100 entries to statistically analyze data 105 on a segment-by-segment basis.
  • a segment may be defined as a sentence and/or heading and/or title.
  • a segment may be defined as a paragraph and/or heading and/or title.
  • a segment may be defined as a chapter and/or heading and/or title.
  • a segment may be defined as a complete document and/or heading and/or title.
  • Other definitions may be appropriate for certain types of data and, while different from those enumerated here, would be obvious to one of ordinary skill in the art.
  • headings and titles may be excluded from consideration. It is noted that only a portion of data 105 need be analyzed in accordance with block 1 lO.That is, a first portion of data 105 may be used to generate topic list 115, with the topics so identified being associated with the entire corpus of data during the acts of block 120. TABLE l
  • data 105 comprises the text of approximately 12 million abstracts from the Medline® data collection. These abstracts include approximately 2.8 million unique words, representing approximately 40 Gigabytes of raw data.
  • MEDLINE® Medical Literature, Analysis, and Retrieval System Online
  • NLM National Library of Medicine's
  • the database contains bibliographic citations and author abstracts from more than 4,600 biomedical journals published in the United States and 71) other countries.
  • Medline M is searchable at no cost from the NLM's web site at http://www.nlm.nih.gov.
  • word list 100 may be generated by first compiling a preliminary list of domain specific words 200 and then pruning from that list those entries that do not significantly and,(r uniquely identify concepts or topics within the target domain (block 205).
  • Preliminary list 200 may, for example, be comprised of words from a dictionary, thesaurus, glossary, domain specific word list or a combination of these sources.
  • the Internet may be used to obtain preliminary word lists for virtually any field.
  • Words removed in accordance with block 205 may include standard STOP words as illustrated in Table 2. (One of ordinary skill in the art will recognize that other STOP words may be used.)
  • a general domain word list may be created that comprises those words commonly used in English (or another language), including those that are specific to a number of different domains. This "general word list” may be used to prune words from a preliminary domain specific word list. In another embodiment, some common words removed as a result of the general word list pruning just described may be added back into preliminary word list 200 because, while used across a number of domains, have a particular importance in the particular domain.
  • Example Stop Words a, about affect, after, again, all, along, also, although, among, an, and, another, any, anything, are, as, at, be, became, because, been, before, both, but, by, can, difference, each, even, ever, every, everyone, for, from, great, had, has.
  • preliminary word list 200 was derived from the Unified Medical language System Semantic Network (see http:/www.nlm.nih.gov/datebases/leased.html#umls) and included 4,000,000 unique single- word entries. Of these, roughly 3,945,000 were moved in accordance with block 205. Accordingly, word list 100 comprised approximately 55,000 one word entries.
  • Example word list 200 entries for the medical domain include: abdomen, biotherapy, chlorided, distichiasis, enzyme, enzymes, freckle, gustatory, immune, kyphoplasty, laryngectomy, malabsorption, nebulize, obstetrics, pancytopenia, quadriparesis, retinae, sideeffect, tonsils, unguium!, Vermicular, womb, xerostomia, yersinia, and zygote.
  • word list 100 provides an initial estimation of domain specific concepts/topics. Analysis in accordance with the invention beneficially expands the semantic breadth of word list 100, however, by identifying word collections (e.g., pairs and triplets) as topics (i.e., topic list 115). Once topics are identified, each segment in data 105 may be associated with those topics (block 120) that exist in that segment. Accordingly, if a corpus of data comprises information from a plurality of domains, analysis in accordance with FIG. 1 may be run multiple times-each time with a different word list 100.
  • word collections e.g., pairs and triplets
  • FIG. 3 illustrates one method in accordance with the invention to identify topics (block 110 of FIG.
  • preliminary topic fist 305 A result of this initial step is preliminary topic fist 305.
  • an expected value for each entry in preliminary topic list 305 is computed (block 310) and compared with the actual usage value determined during block 300 (block 315). If the measured actual usage of a preliminary topic list entry is significantly greater than the computed expected value of the entry (the "yes" prong of block 315). that entry is added to topic list 115 (block 320).
  • topic list 115 For the data set identified in Tables 1 and 3, 10 of the 35 Gigabytes were used to generate topic list 115.
  • topic list 115 comprised approximately 506,000 entries. In one embodiment, each of these entries are double word entries.
  • Illustrative topics identified for Medline (9 abstract content in accordance with the invention include: adenine nucleotide, heart disease, left ventricular, atria ventricles, heart failure, muscle, heart rate, fatty acids, loss bone, patient case, bone marrow, and arterial hypertension.
  • one method to measure the actual usage of significant words in data 105 is to determine three statistics for each entry in word list 100: Sl (block 400); S2 (block 405); and S3 (block 410).
  • Sl block 400
  • S2 block 405
  • S3 block 410
  • statistics Sl, S2 and S3 measure the actual frequency of usage of various words and word combinations in data 105 at the granularity of the user-defined segment. More specifically:
  • Statistic Sl (block 400) is a segment-level frequency count for each entry in word list 100.
  • Sl for word-i is the number of unique paragraphs in data 105 in which word-i is found.
  • An Sl value may also be computed for non-word list 100 words if they are identified as part of a word combination as described below with respect to statistic S2.
  • Statistic S2 (block 405) is a segment-level frequency count for each significant word combination in data 105. Those word combinations having a non-zero S2 value may be identified as preliminary topics 305.
  • a "significant word combination" comprises any two entries in word list 100 that are in the same segment.
  • a "significant word combination” comprises any two entries in word list 100 that are in the same segment and contiguous.
  • a "significant word combination” comprises any two entries in word list 100 that are in the same segment and contiguous or separated only by one or more STOP words.
  • a "significant word combination” comprises any two words that are in the same segment and contiguous or separated only by one or more STOP words where at least one of the words in the word combination is in word list 100.
  • a "significant word combination” comprises a two or more word combination appearing in any data item within Data 105.
  • word list 100 would not be used.
  • a "significant word combination” comprises any two or more words that are in the same segment and separated by 1 N 1 or fewer specified other words: N may be zero or more; and the specified words are typically STOP words.
  • word combinations comprising non-word list 100 words may be ignored if they appear in less than a specified number of segments in data 105 (e.g., less than 10 segments).
  • S2 for word-combination-i is the number of unique paragraphs in data 105 in which word-combination-i is found.
  • Statistic S3 (block 410) indicates the number of unique word combinations (identified by having non-zero S2 values, for example) each word in word list 100 was found in.
  • word-z's S3 value is 3.
  • One method to compute the expected usage of significant words in data 105 is to calculate the expected value for each preliminary topic list 305 entry based only on its overall frequency of use in data 105.
  • the expected value .for each word pair in preliminary word list 305 may be computed as follows:
  • Sl(word-i) and Sl(word-j) represents the Sl statistic value for word-i and word-j respectively
  • N represents the total number of segments in the data corpus being analyzed.
  • the test (block 315) of whether a topic's measured usage (block 300) is significantly greater than the topic's expected usage (block 310), is a constant multiplier. For example, if the measured usage of preliminary topic list entry-i is twice that of preliminary topic list entry-i is expected usage, preliminary topic list entry-i may be added to topic list 115 in accordance with block 320. In another embodiment of the invention, if the measured usage of preliminary topic list entry-i is greater than a threshold value (e.g., 10) across all segments, then that preliminary topic list entry is selected as a topic.
  • a threshold value e.g. 10
  • a different multiplier may be used (e.g., 1.5 or 3). Additionally conventional statistical tests of significance may be used.
  • topic list 115 may be refined in accordance with
  • FIG. 5 (For convenience, this refinement process will be described in terms of two-word topics. One of ordinary skill in the art will recognize that the technique is equally applicable to topics having more than two words.)
  • a first two word topic is selected (block 500). If both words comprising the topic are found in word list 100 (the "Yes" prong of block 505), the two word topic is retained (block 510). If both words comprising the topic are not found in word list 100 (the "no" prong of block 505), but the S3 value for that word which is in word list 100 is not significantly less than the S3 value for the other word (the yes" prong of block 515), the two word topic is retained (block 510).
  • the test for significance is based on whether the "high" S3 value is in the upper one-third of all S3 values and the "low" S3 value is in the lower one-third of all S3 values.
  • the test for significance in accordance with block 515 may be based on quartiles, quintiles or Bayesian tests. Refinement processes such as that outlined in FIG. 5 acknowledge word associations within data, while ignoring individual words that are so prevalent alone (high S3 value) as to offer substantially no differentiation as to content.
  • each segment in data 105 may associated with those topics which exist within it (block 120) and stored in database 125.
  • Topics may be associated with a data segment in any desired fashion. For example, topics found in a segment may be stored as metadata for the segment. In addition, stored topics may be indexed for improved retrieval performance during subsequent lookup operations.
  • Empirical studies show that the large majority of user queries are "under-defined.” That is, the query itself does not identify any particular subject matter with sufficient specificity to allow a search engine to return the user's desired data in a result set (i.e., that collection of results presented to the user) that is acceptably small.
  • a typical user query may be a single word such as, for example, "kidney.”
  • prior art search techniques generally return large result sets—often containing thousands, or tens of thousands, of "hits.” Such large result sets are almost never useful to a user as they do not have the time to go through every entry to find that one having the information they seek.
  • topics associated with data segments in accordance with the invention may be used to facilitate data retrieval operations as shown in FIG. 6.
  • a user query When a user query is received (block 600) it may be used to generate an initial result set (block 605) in a conventional manner. For example, a literal text search of the query term may identify 100,000 documents (or objects stored in database 125) that contain the search term. From this initial result set, a subset may be selected for analysis in accordance with topics (block 610). In one embodiment, the subset is a randomly chosen 1% of the initial result set. In another embodiment, the subset is a randomly chosen 1,000 entries from the initial result set. In yet another embodiment, a specified number of entries are selected from the initial result set (chosen in any manner desired).
  • the number of entries in the initial result subset may be chosen in substantially any manner desired, it is preferable to select at least a number that provides "coverage" (in a statistical sense) for the initial result set. In other words, it is desirable that the selected subset mirror the initial result set in terms of topics. With an appropriately chosen result subset, the most relevant topics associated with those results may be identified (block 615) and displayed to the user (block 620).
  • FIG. 7 shows one method in accordance with the invention to identify those topics for display (block 615). Initially, all unique topics associated with the result subset are identified (block 700), and those topics that appear in more than a specified fraction of the result subset are removed (block 705). For example, those topics appearing in 80% or more of the segments comprising the result subset may be ignored for the purposes of this analysis. (A percentage higher or lower than this may be selected without altering the salient characteristics of the process.) Next, that topic which appears in the most result subset entries is selected for display (block 710). If more than one topic ties for having the most coverage, one may be selected for display in any manner desired.
  • the specified threshold of block 715 is 20%, although a percentage higher or lower than this may be selected without altering the salient characteristics of the process.
  • the remaining topics are serialized and duplicate words are eliminated (block 725). That is, topics comprising two or more words are broken apart and treated as single-word topics.
  • the single- word topic that appears in the most result subset entries not already excluded is selected for display (block 730). As before, if more than one topic ties for having the most coverage, one may be selected for display in any manner desired.
  • the topics identified in accordance with FIG. 7 may be displayed to the user (block 620 in FIG. 6).
  • data retrieval operations in accordance with the invention return one or more topics which the user may select to pursue or redefine their initial search.
  • a specified number of search result entries may be displayed in conjunction with the displayed topics.
  • a user may be presented with those data corresponding to the selected topics.
  • Topicics may, for example, be combined through Boolean "and" and/or "or” operators.
  • the user may be presented with another list of topics based on the "new" result set in a manner described above.
  • search operations in accordance with the invention respond to user queries by presenting a series of likely topics that most closely reflect the subjects that their initial search query relate to. Subsequent selection of a topic by the user, in effect, supplies additional search information which is used to refine the Search.
  • a search on the single word "kidney” returns an initial result set comprising 147,549 hits. (That is, 147,549 segments had the word kidney in them.) Of these, 1,000 were chosen as the initial result subset.
  • the following topics were represented in the result set: amino acid, dependent presence, amino terminal, kidney transplantation, transcriptional regulation, liver kidney, body weight, rat kidney, filtration fraction, rats treated, heart kidney, renal transplantation, blood pressure, and renal function.
  • renal function topic identified a total of 6,853 entries divided among the following topics: effects renal, kidney transplantation, renal parenchyma, glomerular filtration, loss renal, blood flow, histological examination, renal artery, creatinine clearance, intensive care, and renal failure. Selection of the "glomerular filtration” topic from this list identified a total of 1,400 entries. Thus, in two steps the number of "hits" through which a person must search was reduced from approximately 148,000 to 1,500-a reduction of nearly two orders of magnitude.
  • retrieval operations in accordance with FIG. 6 may not be needed for all queries. For example, if a user query includes multiple search words or a quoted phrase that, using literal text-based search techniques, returns a relatively small result set (e.g., 50 hits or fewer), the presentation of this relatively small result set may be made immediately without resort to the topic-based approach of FIG. 6. What size of initial result set that triggers use of a topic-based retrieval operation in accordance with the invention is a matter of design choice. In one embodiment, all initial result sets having more than 50 hits use a method in accordance with FIG. 6. In another embodiment, only initial result sets having more than 200 results trigger use of a method in accordance with FIG. 6.
  • FIG. 8 provides a diagram that shows enterprise information sources.
  • Enterprise information can include data warehouses, multiple databases, and document systems.
  • Server and PC information can include reports, presentations and data generated by the worker or his colleagues.
  • Internet information can include a wealth of information, including business websites and business news.
  • the first step in addressing the information dilemma is to provide realtime aggregation of information where the context (e.g. title, to, from, name, product, etc.) is identified and maintained. This must be done without requiring normalization of the data. Or, in other words, the information must be imported "as is” without having to reformat or transform the information into some common form. Examples of methods for aggregating the data are taught in commonly owned U.S. patent number 5,842,213, entitled Method for Modeling, Storing and Transferring Data in Neutral Form, issued Nov. 24, 1998 to Odom et al., and U.S.
  • the second step relates to the search problem, or put another way, finding the needed information - the proverbial needle in the haystack.
  • search engines do not display topics associated with every subject matter domain related to a search constraint entered by a user. Rather a search engine may only show search results that are most popular without regard to different subject matter domains that search results may belong to. For users interested in a particular domain, the search results displayed would not be particularly relevant and their specific areas of interest difficult to find. Thus, a user once again may not receive search results relevant to their particular area of interest.
  • the present invention addresses these shortcomings of existing search engines and methods.
  • embodiments of the present invention provides search methods and systems that can efficiently generate search results to identify and display topics by considering, at any given time, the relative significance of a topic based on current events and that ensure coverage of all subject matter domains associated with a search constraint.
  • a topic comprise a word combination of two or more substantially contiguous words. Two words are substantially contiguous if they are separated only by zero or more words selected from a predetermined list of words. In one embodiment, the predetermined list of words are STOP words.
  • the set of information includes one or more of information located within an enterprise network, information located within a server, information located within a personal computer, information located on the Internet, or information contained within email messages or email attachments.
  • data item includes one or more of text documents, graphic documents, audio files, video files, multimedia documents, email messages, email attachments, or Internet web page.
  • FIG. 9 provides a flowchart of method 900 for displaying topics related to a search constraint entered by a user to request search results that identify data items within a set of information that are related to the search constraint, according to an embodiment of the invention.
  • Method 900 begins in step 910.
  • FIG. 10 provides a screen shot of an search engine web site, according to an embodiment of the invention. The screen shot of FIG. 10 is for illustrative purposes, and not intended to limit the scope of the invention.
  • step 910 a search constraint is received. For example, referring to
  • FIG. 10 the search constraint is "Pittsburgh Steelers.”
  • a first preliminary set of topics related to the search constraint is identified.
  • the first preliminary set of topics is representative of a sample set of general data items.
  • the general data items could include a generic sampling of data items located across the Internet.
  • a second preliminary set of topics related to the search constraint is identified.
  • the second preliminary set of topics are representative of a sample set of current event data items.
  • the sample set of current event data items are gathered by receiving feeds from current event websites, such as CNN.COM, MSN.COM, ESPN.COM and the like.
  • the current event data items are updated periodically. In one embodiment periodic updates are a function of the subject matter. For example, sports information is updated every thirty minutes, financial information is updated every thirty minutes, health information is updated once a day and other news information is updated every two hours.
  • the current event data items database contains approximately 20,000 data items.
  • a set of display topics is identified that is a subset of the first preliminary set of topics and the second preliminary set of topics.
  • identifying a set of display topics includes selecting a certain number, referred to as the general topic threshold number, of topics from the first preliminary set of topics and selecting a certain number, referred to as the current event topic threshold number of topics, from the second preliminary set of topics. Additionally, in a further embodiment a certain number, referred to as the proper name topic threshold, of proper names from the second preliminary set of topics are also selected. In one embodiment, the proper names are randomly selected from a set of proper names contained within the second preliminary set of topics.
  • a personal interest topic repository can be created.
  • the personal interest topic repository includes topics that have been identified as relevant to a user. These topics, for example, may be topics associated with frequent searches conducted by a user, topics generated based on a personal profile, or topics that a user may have previously selected.
  • step 940 can also include selecting a certain number, referred to as the personal interest topic threshold, of topics from the first preliminary set of topics.
  • step 950 the set of display topics identified in step 940 is displayed.
  • the topics may be displayed on a computer terminal, cell phone or other display device.
  • step 960 method 900 ends.
  • the topic display threshold is twenty topics. Of these twenty topics, six topics are identified from the current event topics, six proper names (which are considered topics) are also taken from the current event topics, and eight topics are identified from the general topics. Of the eight topics from the general topics, two of these are personal interest topics, when personal interest topics are available.
  • the column labeled AUTOTOPICS displays the set of display topics.
  • the topics include, for example, Franco Harris, Pittsburgh Post, and autographed photos.
  • FIG. 11 provides a flowchart of method 1100 for displaying topics related to a search constraint entered by a user to request search results that identify data items within a set of information that are related to the search constraint. For use in illustrating the steps in method 1100, FIG. 10 will again be used.
  • the screen shot of FIG. 10 is for illustrative purposes, and not intended to limit the scope of the invention.
  • Method 1100 begins in step 1110.
  • step 1110 a search constraint is received. For example, referring to
  • FIG. 10 the search constraint is "Pittsburgh Steelers.”
  • step 1120 a set of topics related to the search constraint is identified.
  • identifying a set of topics includes conducting a search to generate search results.
  • the search results include a set of data items.
  • Example searches that can be used include searches using GOOGLE, YAHOO, MSN, ASK.COM and A9 search engines. Other types of search engines can also be used.
  • a search can be conducted on a representative sample of data within the set of information that is of interest. For example, when searching the Internet a representative set of data items from the Internet can be used. In one embodiment the representative set of data items includes about 25 million data items.
  • a search can be conducted on data items contained within a current events data item database.
  • the sample set of current event data items are gathered by receiving feeds from current event websites, such as CNN.COM, MSN.COM, ESPN.COM and the like.
  • the current event data items are updated periodically.
  • periodic updates are a function of the subject matter. For example, sports information is updated every thirty minutes, financial information is updated every thirty minutes, health information is updated once a day and other news information is updated every two hours.
  • the current event data items database contains approximately 20,000 current event data items.
  • the set of topics can then be determined from the search results by extracting topics associated with each data item in the search results.
  • topification methods disclosed in the "026 Patent Application can be used to identify the set of topics from any of the above search results using general data items, representative data items and current event data items.
  • topics can be generated from a combination of these or other source data items.
  • FIG. 13 provides a flowchart of a method 1300 to rank topics into one of four, general rankings.
  • the highest ranking is assigned to a topic when the topic is a current topic and a personal interest topic.
  • a topic is a current topic when it is found in the current event topics.
  • a topic is a personal interest topic when it is found in the personal interest topic repository for a particular user.
  • step 1320 the second highest ranking is assigned to a topic within the identified when the topic is a current topic.
  • step 1330 the third highest ranking is assigned to a topic when the topic is a personal interest topic.
  • step 1340 the fourth highest ranking is assigned to a topic when the topic is neither a current topic or a personal interest topic.
  • topics are further ranked based on their frequency of occurrence with search result data items. Those topics that occur least frequently among the data items are considered most relevant and given a higher ranking.
  • FIG. 14 provides a diagram that graphically illustrates this process.
  • Set 1410 represents the complete set of topics found in the data items in the search results.
  • three subject matter domains are illustrated. These are subject matter domains 1420, 1430 and 1440.
  • Subject matter domains include a collection of topics associated with the data items within the search results.
  • subject matter domains includes data item 1450.
  • Associated with data item 1450 will be one or more topics.
  • Data items that have overlapping sets of topics, represented by the shaded area 1460 for subject matter domain 1430, are clustered together to form a subject matter domain.
  • Subject matter domains will have some overlap, as indicated by overlap 1470.
  • the process of clustering includes clustering data items that have overlapping topics, and then creating subject matter domains based on clustering of data items that minimizes the overlap of topics across subject matter areas, such as overlap 1470.
  • Individuals skilled in the relevant arts will be able to apply statistical clustering methods to determine the optimal clustering.
  • step 1150 the most representative topic for each subject matter domain is determined.
  • the most representative topic is determined by identifying those topics within a subject matter domain that occur in more than some fraction of the distribution (e.g, more than 90% of the data items) of data items within the set of information.
  • the most representative topic is then determined from this set of topics by identifying the topic for each subject matter domain with the highest current event and personal interest topic ranking. As necessary, the frequency of occurrence of the topics can be used to further rank the topics as discussed above.
  • step 1160 the most representative topic for each subject matter domain is displayed.
  • step 1170 the highest ranked topics not previously displayed are displayed.
  • step 1180 method 1100 ends.
  • FIG. 12 provides a flowchart of method 1200 for displaying topics related to a search constraint entered by a user to request search results that identify data items within a set of information that are related to the search constraint.
  • FIG. 10 will again be used.
  • the screen shot of FIG. 10 is for illustrative purposes, and not intended to limit the scope of the invention.
  • Method 1200 begins in step 1210.
  • step 1210 a search constraint is received. For example, referring to
  • FIG. 10 the search constraint is "Pittsburgh Steelers.”
  • step 1220 a set of topics related to the search constraint is identified.
  • identifying a set of topics includes conducting a search to generate search results.
  • the search results include a set of data items.
  • Example searches that can be used include searches using GOOGLE, YAHOO, MSN, ASK.COM and A9 search engines. Other types of search engines can also be used.
  • a search can be conducted on a representative sample of data within the set of information that is of interest. For example, when searching the Internet a representative set of data items from the Internet can be used. In one embodiment the representative set of data items includes about 25 million data items.
  • a search can be conducted on data items contained within a current event data item database.
  • the sample set of current event data items are gathered by receiving feeds from current event websites, such as CNN.COM, MSN.COM, ESPN.COM and the like.
  • the current event data items are updated periodically.
  • periodic updates are a function of the subject matter,. For example, sports information is updated every thirty minutes, financial information is updated every thirty minutes, health information is updated once a day and other news information is updated every two hours.
  • the current event data items database contains approximately 20,000 current event data items.
  • the set of topics can then be determined from the search results by extracting topics associated with each data item in the search results.
  • the topif ⁇ cation methods disclosed in the "026 Patent Application can be used to identify the set of topics from any of the above search results using general data items, representative data items and current event data items.
  • topics can be generated from a combination of these or other source data items.
  • FIG. 14 provides a diagram that graphically illustrates this process.
  • Set 1410 represents the complete set of topics found in the data items in the search results.
  • three subject matter domains are illustrated. These are subject matter domains 1420, 1430 and 1440.
  • Subject matter domains include a collection of topics associated with the data items within the search results.
  • subject matter domains includes data item 1450.
  • Associated with data item 1450 will be one or more topics.
  • Data items that have overlapping sets of topics, represented by the shaded area 1460 for subject matter domain 1430, are clustered together to form a subject matter domain.
  • Subject matter domains will have some overlap, as indicated by overlap 1470.
  • the process of clustering includes clustering data items that have overlapping topics, and then creating subject matter domains based on clustering of data items that minimizes the overlap of topics across subject matter areas, such as overlap 1470.
  • Individuals skilled in the relevant arts will be able to apply statistical clustering methods to determine the optimal clustering.
  • step 1240 the most representative topic for each subject matter domain is determined.
  • the most representative topic is determined by identifying those topics within a subject matter domain that occur in more than some fraction of the distribution of data items (e.g., more than 90% of the data items) within the set of information.
  • the most representative topic is then determined from this set of topics by identifying the topic for each subject matter domain that has the least frequent number of occurrences in the search result data items.
  • step 1250 the most representative topic for each subject matter domain is displayed.
  • step 1250 method 1200 ends.
  • the set of topics identified that are related to the search constraint can be ranked as was done in step 1130 in method 1100. Based on these rankings, additional topics can be displayed as was done in step 1170 in method 1100.
  • FIG. 15 acts in accordance with any, or a portion of any, of FIGS. 1 through 14 may be performed by a programmable control device executing instructions organized into one or more program modules 1500.
  • a programmable control device can include, but is not limited to a personal computer, a laptop computer, a network computer, a wireless telephone, a personal data assistant ("PDA") and the like.
  • PDA personal data assistant
  • programmable control device comprises computer system 1505 that includes central processing unit .1510, storage 1515, network interface card 1520 for coupling computer system 1505 to network 1525, display unit 1530, keyboard 1535 and mouse 1540.
  • a programmable control device may be a multiprocessor computer system or a custom designed state machine.
  • Custom designed state machines may be embodied in a hardware device such as a printed circuit board comprising, discrete logic, integrated circuits, or specially designed Application Specific Integrated Circuits (ASICs).
  • Storage devices such as device 1515, suitable for tangibly embodying program module(s) 1500 include all forms of non-volatile memory including, but not limited to: semiconductor memory devices such as Electrically Programmable Read Only Memory (EPROM), Electrically Erasable Programmable Read Only Memory (EEPROM), and flash devices; magnetic disks (fixed, floppy, and removable); other magnetic media such as tape; and optical media such as CD-ROM disks.
  • EPROM Electrically Programmable Read Only Memory
  • EEPROM Electrically Erasable Programmable Read Only Memory
  • flash devices such as magnetic disks (fixed, floppy, and removable); other magnetic media such as tape; and optical media such as CD-ROM disks.

Landscapes

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

Abstract

Procédés et systèmes relatifs aux moteurs de recherche pour l'affichage de sujets pertinents et opportuns. Selon une variante, on décrit un procédé d'affichage de sujets liés à une contrainte de recherche introduite par un utilisateur, qui consiste à recevoir une contrainte de recherche, à identifier une première série préliminaire de sujets liés à cette contrainte, la première série étant représentative d'une série échantillon d'éléments de données généraux; à identifier une seconde série préliminaire de sujets liés à cette contrainte, la seconde série étant représentative d'une série échantillon d'éléments de données propres à des événements courants; à identifier une série de sujets à afficher qui est un sous-ensemble de la première série et de la seconde série; et à afficher la série de sujets à afficher. Sous d'autres variantes, on décrit d'autres systèmes et procédés.
PCT/US2007/005222 2006-03-01 2007-03-01 Procédés et systèmes relatifs aux moteurs de recherche pour l'affichage de sujets pertinents WO2007103096A2 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2008557356A JP2009528630A (ja) 2006-03-01 2007-03-01 関連するトピックを表示するサーチ・エンジンの方法及びシステム

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US77757606P 2006-03-01 2006-03-01
US60/777,576 2006-03-01

Publications (2)

Publication Number Publication Date
WO2007103096A2 true WO2007103096A2 (fr) 2007-09-13
WO2007103096A3 WO2007103096A3 (fr) 2008-04-17

Family

ID=38475378

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2007/005222 WO2007103096A2 (fr) 2006-03-01 2007-03-01 Procédés et systèmes relatifs aux moteurs de recherche pour l'affichage de sujets pertinents

Country Status (2)

Country Link
JP (1) JP2009528630A (fr)
WO (1) WO2007103096A2 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108701133A (zh) * 2016-11-30 2018-10-23 微软技术许可有限责任公司 提供推荐内容

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030167252A1 (en) * 2002-02-26 2003-09-04 Pliant Technologies, Inc. Topic identification and use thereof in information retrieval systems
US20060004732A1 (en) * 2002-02-26 2006-01-05 Odom Paul S Search engine methods and systems for generating relevant search results and advertisements

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003345810A (ja) * 2002-05-28 2003-12-05 Hitachi Ltd 文書検索方法、文書検索システム及び文書検索結果示方システム
US7447678B2 (en) * 2003-12-31 2008-11-04 Google Inc. Interface for a universal search engine

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030167252A1 (en) * 2002-02-26 2003-09-04 Pliant Technologies, Inc. Topic identification and use thereof in information retrieval systems
US20060004732A1 (en) * 2002-02-26 2006-01-05 Odom Paul S Search engine methods and systems for generating relevant search results and advertisements

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108701133A (zh) * 2016-11-30 2018-10-23 微软技术许可有限责任公司 提供推荐内容
US11494450B2 (en) 2016-11-30 2022-11-08 Microsoft Technology Licensing, Llc Providing recommended contents

Also Published As

Publication number Publication date
WO2007103096A3 (fr) 2008-04-17
JP2009528630A (ja) 2009-08-06

Similar Documents

Publication Publication Date Title
US7716207B2 (en) Search engine methods and systems for displaying relevant topics
US20060004732A1 (en) Search engine methods and systems for generating relevant search results and advertisements
US7340466B2 (en) Topic identification and use thereof in information retrieval systems
Tsatsaronis et al. Bioasq: A challenge on large-scale biomedical semantic indexing and question answering
US8521734B2 (en) Search engine with augmented relevance ranking by community participation
US8060513B2 (en) Information processing with integrated semantic contexts
Kalashnikov et al. Web people search via connection analysis
US20100005087A1 (en) Facilitating collaborative searching using semantic contexts associated with information
US20070185847A1 (en) Methods and apparatus for filtering search results
Trillo et al. Using semantic techniques to access web data
Wolfram Search characteristics in different types of Web-based IR environments: Are they the same?
AU2011210742A1 (en) Method and system for conducting legal research using clustering analytics
Lee et al. A social inverted index for social-tagging-based information retrieval
Sun et al. A novel approach for personalized article recommendation in online scientific communities
Price et al. Using semantic components to search for domain-specific documents: An evaluation from the system perspective and the user perspective
Spangler et al. Simple: Interactive analytics on patent data
WO2007103096A2 (fr) Procédés et systèmes relatifs aux moteurs de recherche pour l'affichage de sujets pertinents
Gan et al. A query transformation framework for automated structured query construction in structured retrieval environment
EP1836555A2 (fr) Procedes et systemes de moteurs de recherche permettant de generer des resultats de recherche et des annonces publicitaires pertinents
Toda et al. Search result clustering using informatively named entities
Kanavos et al. On topic categorization of pubmed query results
Bilal University Mohamed Boudiaf of M’sila
Briscoe et al. Intelligent information access from scientific papers
Heenan A Review of Academic Research on Information Retrieval
Jagan et al. A Query Recommendation System for Efficient Biomedical Information Retrieval

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 2008557356

Country of ref document: JP

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 07751953

Country of ref document: EP

Kind code of ref document: A2