Abstract: A method, system and computer program product is disclosed for intelligent data mining. The method comprises receiving an event from an application, assigning property weights to properties of the event, and building a query from these properties based on the property weights. The method further comprises assigning search engine weights to a group of search engines, selecting at least some of the search engines based on the search engine weights, and sending the built query to the selected search engines. Results from the selected search engines are stored in a knowledge repository and used to adjust the property weights and the search engine weights. The invention may be used to provide an analysis with information about a problem, and to manage a solutions database which can be used for problem determination. The invention provides a low cost solution for collecting relevant information from online sources.

Abstract: Described herein are methods and systems for pattern recognition in web search engine result pages. The input data is a result page from a web search engine as well as an integer number for the results on the page. The output is a regular expression that matches all the results on the page, capturing each result and its individual fields.

Abstract: Described are a system and method for determining an event occurrence rate. A sample set of content items may be obtained. Each of the content items may be associated with at least one region in a hierarchical data structure. A first impression volume may be determined for the at least one region as a function of a number of impressions registered for the content items associated with the at least one region. A scale factor may be applied to the first impression volume to generate a second impression volume. The scale factor may be selected so that the second impression volume is within a predefined range of a third impression volume. A click-through-rate (CTR) may be estimated as a function of the second impression volume and a number of clicks on the content item.

Abstract: A network's evolution is characterized by graph evolution rules. A graph that represents an evolutionary network is mined to identify evolutional patterns of the network, and graph evolution rules are generated using identified evolutional patterns. The generated graph evolution rules represent the evolutional patterns of the network.

Abstract: Techniques and tools described herein mine social information from a source and store the social information in a database. Responsive to a search object, the techniques search the stored social information and determine social relationships. The techniques further provide, via a graphical user interface, the social relationships determined from the social information stored in the database. In several embodiments, the techniques enable social relationship feedback.

Abstract: Various embodiments disclosed herein are directed to managing and sharing data between web accessed calculators. The systems include a data store to persist calculator inputs and outputs and share them with other calculators and with customer service representatives.

Abstract: Described herein is a technology that facilitates learning interests for advertising based on automated analysis of images. In several embodiments a person's interests are automatically learned based on the person's photographs for targeted advertising. Techniques are described that facilitate automatically detecting a user's interest from images and suggesting user-targeted ads. As described herein, these techniques include computer-annotating images with learned tags, performing topic learning to obtain an interest model, and performing advertisement matching and ranking based on the interest model.

Abstract: A computer system and method is disclosed for mining current and archived address data in order to identify a preferred address for each service point in a territory. The data mining system may start in response to the presentation of a candidate address for matching. The set of mined data may be prioritized by clustering like characteristics, building similarity matrices, and by constructing dendrograms with nodes joined according to common characteristics. A computer system and method for maintaining a central database of preferred addresses is also disclosed. Selected address data gathered in a queue may be scored by characteristic, grouped by consignee location, and staged for processing. The scored queue of data may be prioritized by clustering like characteristics, building similarity matrices, and by constructing dendrograms.

Abstract: Described is a technology for automatically generating labeled training data for training a classifier based upon implicit information associated with the data. For example, whether a query has commercial intent can be classified based upon whether the query was submitted at a commercial website's search portal, as logged in a toolbar log. Positive candidate query-related data is extracted from the toolbar log based upon the associated implicit information. A click log is processed to obtain negative query-related data. The labeled training data is automatically generated by separating at least some of the positive candidate query data from the remaining positive candidate query data based upon the negative query data. The labeled training data may be used to train a classifier, such as to classify an online search query as having a certain type of intent or not.

Abstract: The present disclosure provides a method and apparatus of knowledge base building to automatically construct a knowledge base. Furthermore, the disclosed techniques can be used to improve the accuracy of that knowledge base. In one aspect, a method acquires a sentence from a webpage using a basic data processing layer of a computing apparatus. The acquired sentence is parsed into words using a data mining layer of the computing apparatus. One or more representative words in a first category of a knowledge base are matched with the words parsed from the acquired sentence. When there is a match between one of the representative words and one of the words parsed from the acquired sentence, a string of words adjacent the matched word in the acquired sentence is added to the first category as a first entry.

Abstract: Privacy in data mining of sparse high dimensional data records is preserved by transforming the data records into anonymized data records. This transformation involves creating a sketch-based private representation of each data record, each data record containing only a small number of non-zero attribute value in relation to the high dimensionality of the data records.

Abstract: Data mining for organization insights may be provided. Data from a plurality of sources, such as user communications and documents, may be collected. The collected data may be analyzed to identify an insight about users or organizations associated with the communications. The insight may be provided to a user, such as in response to a search query, an analytics tool, or an added application functionality.

Abstract: A method and computer system for reporting on a target greenhouse gas within a geographical boundary of an offset project by compiling policy parameters for the target greenhouse gas and generating a science plan for monitoring the target greenhouse gas for the target geographical boundary of the offset project, based upon the compiled policy parameters. An allometric model for the target greenhouse gas within the geographical boundary of the offset project is generated based upon the science plan of the target greenhouse gas for the geographic boundary, and a report for the target greenhouse gas within the target geographical boundary of the offset project is generated based upon the allometric model.

Abstract: When treating a patient, clinical decision support system (CDSS) guidelines are employed to assist a physician in generating a treatment plan. These plans are generated using both imaging and non-imaging data. To accomplish this, the CDSS is interfaced with imaging systems (CADx, CAD, PACS etc.). A data-mining operation is performed to identify relevant patients with similar attributes such as diagnosis, medical history, treatment, etc from imaging and non-imaging data. Natural language processing is employed to extract and encode relevant non-imaging (textual) data from relevant patients' records. Additionally, an image of a current patient is compared to reference images in a patient database to identify relevant patients. Relevant patients are then identified to a user, and the user selects a relevant patient to view detailed information related to medical history, treatment, guidelines, efficacy, and the like.

Abstract: A computer processing device receives computer readable data to derive computer executable rules for mining and constructing situation categories. The received data is transformed into a predetermined standard format if the received data is not already in the predetermined standard format. The predetermined standard formatted data is parsed, and an outer, iterative loop is performed until at least one predetermined stopping criterion is met. An inner iterative loop is performed within the outer iterative loop until all desired subsets of data are processed. During the inner iterative loop, selected subsets of data are labeled with labels associated with corresponding previously labeled subsets of data. New computer executable rules are generated for mining and constructing situation categories from the labeled subsets of data. Keyword list classifiers are transformed using the stored labeled subsets of data.

Abstract: A system and method for summarizing jobs for a user group is provided. In one embodiment, a job manager is operable to invoke an alert filter. The alert filter is compatible with a plurality operating environments. One or more properties of a first job associated with a first operating environment is identified. One or more properties of a second job associated with a second operating environment is identified. The first operating environment and the second operating environment are heterogeneous. A first alert object is generated in response to a first match between the alert filter and the identified properties of the first job. A second alert object is generated in response to a second match between the alert filter and the identified properties of the second job.

Abstract: A computer automated method of aggregating and presenting data includes the steps of inputting a set of user-defined instructions into a computer database system, inputting a user query into the computer database system, mining the computer database system for data relevant to the user query, creating a data set comprising said data relevant to the user query, and aggregating data in the data set using domain metrics selected based on any of predefined and configurable rules and past user usage, selecting at least one presentation report for compiling the aggregated data, wherein the selection is based on any of predefined and configurable rules and past user usage, and displaying the at least one presentation report to the user, wherein the displaying process comprises graphically arranging the at least one presentation report based on an available viewing area of a device accessing the at least one presentation report.

Abstract: A method and system is provided for managing business taxonomy. The system comprises an indexing engine for indexing content of source business oriented metadata. The indexing engine has a content scanner for reading the business oriented metadata, defining taxonomy of the business oriented metadata, and building a content index of the business oriented metadata including a subject index representing the taxonomy of the business oriented metadata. The system also comprises an index store for storing the content index of the business oriented metadata, and a taxonomy engine for providing taxonomy services to users using the content index.

Abstract: Techniques for using metadata as comments to assist with search problem determination and analysis are provided. Before an action is taken on a search, contextual information is gathered as metadata about the action and actor requesting the action. The metadata is embedded in the search as comments and the comments are subsequently logged when the action is performed on the search. The comments combine with other comments previously recorded to permit subsequent analysis on searches.

Abstract: Methods and apparatus, including computer program products, implementing and using techniques for pattern detection in input data containing several transactions, each transaction having at least one item. Filter conditions for interesting patterns are received, and a first set of filter conditions applicable in connection with generation of candidate patterns is determined. An evaluated candidate pattern is selected as a parent candidate pattern, and evaluation information about the parent candidate pattern is maintained. Child candidate patterns are generated by extending the parent candidate pattern and taking into account the first set of filter conditions. The child candidate patterns are evaluated with respect to the input data together in sets of similar candidate patterns and based on the evaluation information about the parent candidate pattern. At least one child candidate pattern successfully passing the evaluation step is recursively used as a parent candidate pattern.

Abstract: A system, method, and computer program product provides a multi-category apply operation in a data mining system that produces output with multiple class values, their associated measures including probabilities in case of supervised models, quality of fit and distance in case of clustering models, and the relative ranks of the predictions. A method for multi-category apply in a data mining system comprises the steps of receiving input data for scoring including a plurality of rows of data applied to a data mining model and generating multi-category apply output including a plurality of class values and their associated probabilities based on the received input data, the selected class values having probabilities meeting a selection criterion and their ranks.

Abstract: Provided is an image search device which relatively easily searches a large amount of stored images for images that a user wishes to use for interpolation, and which includes: an interpolation range computing unit (103) which computes, as an interpolation range, a range of an area including a first photographing location where a first interpolation target image was taken and a second photographing location where a second interpolation target image was taken; an interpolation image candidate obtaining unit (104) which obtains, as candidate images, images whose photographing location are included in the interpolation range from the plurality of images; and an interpolation image selecting unit (106) which selects, from the candidate images, an image having a greater subject distance, which is a distance between a subject and an imaging device when the image was taken, as a traveling speed between the first photographing location and the second photographing location increases.

Abstract: Techniques for object relational mapping in database technologies are described herein. According to one embodiment, in response to a query statement for accessing a relational database, a syntax tree is generated to represent semantic information of the query statement, where the query statement has a boolean parameter and is implemented as an SQL object. A data type of the boolean parameter is predicted based on the semantic information obtained from the syntax tree in view of a structure representing the syntax tree. The boolean parameter is configured to be either a numeric value or a string dependent upon metadata used to map the SQL object to the relational database. Other methods and apparatuses are also described.

Abstract: Provided is a text mining device that performs an analysis properly with respect to a difference between plural related document data. Equipped are an element extracting section 140 that extracts language elements from related two or more document data respectively; a differential processing section 150 that extracts a difference between the document data by comparing the elements between the document data which were extracted by the element extracting means 140; and a statistical processing section 170 that performs statistical processing on the difference extracted by the differential processing section 150.

Abstract: Methods for using scenario solution-related information to generate customized user experiences are provided. Upon receiving a user query, a plurality of results is returned, each result being representative of a scenario solution which may be utilized to address a particular issue relevant to the received query. At the time of authoring, each scenario solution is organized based upon one or more keywords and/or one or more categories (i.e., namespaces). Data associated with a namespace/keyword corresponding to a returned search result may be mined to determine information beyond basic scenario solution search results that may be of interest to the user.

Abstract: Techniques for replicating data between database systems without taking checkpoints are provided. In an embodiment, a capture process restarts. Upon restarting, the capture process reestablishes an association with an apply process. A particular logical time maintained by the apply process is then communicated to the capture process. Upon receiving the particular logical time, the capture process restarts mining from this particular logical time.

Abstract: Methods and apparatus, including computer program products, implementing and using techniques for finding deviations in data. A set of candidate patterns is generated. A set of exception patterns that occur in the data less frequently than expected assuming statistical independence is selected from the set of candidate patterns. Data records that comply with at least one of the exception patterns are processed as exception candidates.

Abstract: A method of updating a file attached to an electronic document can include attaching a file to an electronic document and storing a reference to the attached file. The reference can specify a location from which the attached file was obtained. Responsive to a user input, the attached file can be replaced with a file specified by the reference. The attached file also can be updated from newly specified locations.

Abstract: An automated method for ontology generation is provided. In one embodiment, a user inputs a single clinical term or portion of a clinical term representing an adverse event that a patient has experienced. In response, the system causes a list of conceptually related terms to be generated.

Abstract: Methods and systems for solving a target value search problem using a multi-interval heuristic are presented. The methods and system identity a path, or paths, in a graph, whereby a connection graph is created and range sets are generated for each vertex in the connection graph. Range sets include one or more intervals. Thereafter, a best search is performed to identify a path, or paths, from a starting vertex to a goal vertex having a path value closest to a target value.

Abstract: A system and process for record duplication analysis that relies on a multi-membership Bayesian analysis to determine the probability that records within a data set are matches. The Bayesian calculation may rely on objective data describing the data set as well as subjective assessments of the data set. In addition, a system and process for record duplication analysis may rely on the predetermination of probabilistic patterns, where the system only searches for patterns exceeding a chosen threshold.

Abstract: A method and system for cryptographically indexing, searching for, and retrieving documents is provided. In some embodiments, an encryption system is provided that generates a document index that allows users to retrieve documents by performing encrypted queries for keywords associated with the documents. In some embodiments, each keyword maps to the same number of encrypted document identifiers. In some embodiments, an extractor graph is employed to map an indication of each keyword to a number of buckets storing encrypted document identifiers. In some embodiments, an order-preserving encryption system is provided. The encryption system uses an ordered index that maps encrypted instances of ordered attribute values to documents that are associated with those values. The ordered index enables queries containing query operators that rely on order, such as less than (“<”) or greater than (“>”), to be successfully performed on encrypted attribute values.

Abstract: Methods and systems are provided for determining whether to use a full sort sorting technique or a merge sort sorting technique to sort a partially sorted list or data set. One or more tables may be utilized to allow such a determination to be made with regard to a first partially sorted list based on parameters associated with the list including a data distribution type, a number of data items in the list, and a ratio of sorted items to unsorted items in the list.

Abstract: A method, a system and a computer program product for enabling a customer response speech recognition unit to dynamically receive customer feedback. The customer response speech recognition unit is positioned at a customer location. The speech recognition unit is automatically initialized when one or more spoken words are detected. The response statements of customers are dynamically received by the customer response speech recognition unit at the customer location, in real time. The customer response speech recognition unit determines when the one or more spoken words of the customer response statement are associated with a score in a database. An analysis of the words is performed to generate a score that reflects the evaluation of the subject by the customer. The score is dynamically updated as new evaluations are received, and the score is displayed within graphical user interface (GUI) to be viewed by one or more potential customers.

Abstract: An approach is provided for condition monitoring from log messages and sensor trends based on time semi-intervals. The approach may be applied to machine condition monitoring. Patterns are mined from symbolic interval data that extends previous approaches by allowing semi-intervals and partially ordered patterns. The semi-interval patterns and semi-interval partial order patterns are less restrictive than patterns using Allen's relations. Combinations and adaptations of efficient algorithms from sequential pattern and itemset mining for discovery of semi-interval patterns are described.

Abstract: Provided is a charging method capable of offering a user an incentive to use a particle generation factor determining system. In the particle generation factor determining system including a user interface device 11 through which a user inputs a particle map and a server 13, the server 13 calculates accuracy of each of multiple particle generation factors based on the particle map; the user interface device 11 displays the calculated accuracy or a title of generation-factor-relevant information 27 on each particle generation factor corresponding to this accuracy; the server 13 provides the generation-factor-relevant information 27 to the user interface device 11; a charged fee for providing particle generation-factor-relevant information 27 is determined based on accuracy of a particle generation factor corresponding to the provided generation-factor-relevant information 27.

Abstract: Systems and methods are disclosed for saving and restoring the search state of a pattern-recognition processor. Embodiments include a pattern-recognition processor having a state variable array and a state variable storage array stored in on-chip memory (on-silicon memory with the processor). State variable storage control logic of the pattern-recognition processor may control the saving of state variables from the state variable array to the state variable storage array. The state variable storage control logic may also control restoring of the state variables from the state variable storage array to restore a search state.

Abstract: An initial item is grouped into a cluster defined by a query expression applied to a description of the item. Given the initial item, its associated cluster is accessed, and another item is identified based on the initial item's cluster or from a cluster designated as similar to the initial item's cluster. Once identified, the other item is presented as related to the initial item.

Abstract: Media asset interactive search is described. In embodiments, successive keypad number inputs are received that each correlate to multiple characters of one or more different languages. The successive keypad number inputs form an accumulating search key as they are received. A database is searched for a sequence of characters that correlate to the accumulating search key after each successive keypad number input is received. A list of matching terms can then be generated where the matching terms include the sequence of characters, and the list of matching terms narrows with each successive keypad number input. Additionally, the list of the matching terms can be ordered based on a scoring system.

Abstract: Techniques for error-tolerant autocompletion are described. While displaying characters of an input string as they are inputted by a user, when a character is added to the input string by the user, matching strings may be selected from among a set of candidate strings by determining which of the candidate strings have a prefix whose characters match the characters of the input string within a given edit distance of the input string.

Abstract: An account exchange system is provided by a data aggregation service enabled for gathering data for a subscriber from a data repository of a first financial institution, using account exchange software operating on a server coupled to the data aggregation service. Initiated by a subscriber the account exchange software causes an account to be terminated at the first financial institution and a new account to be opened at a second financial institution, using data from the first financial institution, and processing the data to be compatible with data requirements at the second financial institution.

Abstract: The invention comprises a set of complementary techniques that dramatically improve enterprise search and navigation results. The core of the invention is an expertise or knowledge index, called UseRank that tracks the behavior of website visitors. The expertise-index is designed to focus on the four key discoveries of enterprise attributes: Subject Authority, Work Patterns, Content Freshness, and Group Know-how. The invention produces useful, timely, cross-application, expertise-based search and navigation results. In contrast, traditional Information Retrieval technologies such as inverted index, NLP, or taxonomy tackle the same problem with an opposite set of attributes than what the enterprise needs: Content Population, Word Patterns, Content Existence, and Statistical Trends. Overall, the invention emcompasses Baynote Search—a enhancement over existing IR searches, Baynote Guide—a set of community-driven navigations, and Baynote Insights—aggregated views of visitor interests and trends and content gaps.

Abstract: Apparatus having corresponding methods and computer programs, to detect a pattern in a string, comprises a memory circuit to store W-character segments of the pattern, where each segment comprises a fragment of the pattern; a key circuit to generate W-character keys each including a fragment of the string; a comparison circuit to compare the keys and the segments; where, when a segment matches a key, the comparison circuit indicates an initial match between the pattern and the string; and where, when one of the segments matches only a L-character fragment of one of the keys, wherein L<W, the key circuit generates a new key including the L-character fragment and a K-character fragment of the string including K=W?L consecutive characters from the string that are adjacent to the L matching characters in the string.

Abstract: A computer system that includes a processor and a storage medium. The storage medium stores a database of tables and a calculation engine. Each table includes columns and rows. The tables describe components of a hierarchy in which hierarchical relationships between components of the hierarchy are defined. The tables include a component table and component-specific tables. The components of the hierarchy encompass component types. Each component-specific table encompasses only components of the hierarchy. The calculation engine includes a popup calculator being displayed to a user via a display interface of the computer system upon the calculation engine being executed by the processor. The popup calculator performs, under interactive control by the user via the display interface, an evaluation of a specified function of set of functions and displays a result of the evaluation to the user via the display interface.

Abstract: What is disclosed is a novel system and method for analyzing multi-dimensional cluster data sets to identify clusters of related documents in an electronic document storage system. Digital documents, for which multi-dimensional probabilistic relationships are to be determined, are received and then parsed to identify multi-dimensional count data with at least three dimensions. Multi-dimensional tensors representing the count data and estimated cluster membership probabilities are created. The tensors are then iteratively processed using a first and a complementary second tensor factorization model to refine the cluster definition matrices until a convergence criteria has been satisfied. Likely cluster memberships for the count data are determined based upon the refinements made to the cluster definition matrices by the alternating tensor factorization models.

Abstract: An optimization method for a navigation device includes recording a plurality of coordinate variation data, analyzing the plurality of coordinate variation data to generate an analysis result, generating at least one behavior rule according to the analysis result, and adjusting a navigation result of the navigation device according to the at least one behavior rule.

Abstract: A typed separable mixture model is used to mine associative relationships between sets of objects. Instead of modeling only one type of co-occurrence among the sets of objects, the typed separable mixture model can model multiple different types of co-occurrences among more than two sets of objects, and co-occurrences that exist in different contexts.

Abstract: A mechanism by which rule attributes of varying types and numbers can be stored and searched in an efficient manner is provided by storing attribute values of each rule in a child table of a parent rule table. The child table is normalized and contains a foreign key pointing back to the parent rule table and has attribute-value pairs as table columns of the child table. Each rule is then represented by one row of the parent rule table and one or more corresponding rows of the child rule details table. A variable and unlimited number of attribute dimensions is supported among the rules, and search performance is improved through the use of database indexes on the rule details table attribute columns. Metadata representing the structure of the child rule details table will identify the data attributes for each dimension.

Abstract: Apparatuses and methods to perform pattern matching are presented. In one embodiment, an apparatus comprises a memory to store a first pattern table comprising information indicative of whether a byte of input data matches a pattern and whether to ignore other matches of the pattern occur in remaining bytes of the input data. The apparatus further comprises one-byte match logic coupled to the memory, to determine, based on the information in the first pattern table, a one-byte match event with respect to the input data. The apparatus further comprises a control unit to filter the other matches of the pattern based on the information of the first pattern table.

Abstract: Systems, methods and computer readable media are disclosed for a localized gesture aggregation. In a system where user movement is captured by a capture device to provide gesture input to the system, demographic information regarding users as well as data corresponding to how those users respectively make various gestures is gathered. When a new user begins to use the system, his demographic information is analyzed to determine a most likely way that he will attempt to make or find it easy to make a given gesture. That most likely way is then used to process the new user's gesture input.