Abstract: Systems and methods are provided for updating data in a computer network. An exemplary method includes: receiving a first data from one or more data servers; determining a second data for which a first update information is used to update at least one element of the second data; generating a second update information associated with the second data; generating a third update information by executing an operation that compares the first update information and the second update information; generating a third data by applying the third update information to the second data; allowing a user to select via a display device which of the different update elements to update the second data; generating a fourth update information by determining the selected different update elements to update the second data; and updating the second data using the fourth update information.

Abstract: A back-end application computer server may access a potential metadata entries data store containing a set of potential metadata entries, each entry including at least a data element name and a data element definition. A metadata collection system may be executed to automatically populate a metadata template based on the set of potential metadata entries. The system may update entries in the metadata template using a translation tool and validate the updated entries in the metadata template to ensure that required data elements are present. The system may also certify the validated entries load the set of certified metadata entries, including the certified data element names and certified data element definitions, into an enterprise metadata repository data store. Electronic messages may be exchanged to support at least one interactive user interface display associated with certification of the metadata template.

Abstract: A computer-implemented method for transmitting and/or submitting digital content such as feedback for an entity to a blockchain is disclosed. This may be, for example, the Bitcoin blockchain. The method, which is implemented at a first node, includes: submitting, to the blockchain, a first transaction for committing a first quantity of tokens to transfer to a second node associated with the entity, the first transaction having a first output; generating a first content for submission to the blockchain; and submitting, to the blockchain, a second transaction that returns a second quantity of the committed first quantity of tokens to an address associated with the first node, the second transaction having the first output of the first transaction as an input, wherein the first content is used to unlock the first output of the first transaction.

Abstract: An example operation may include one or more of identifying a plurality of instances of a blockchain storage request that have been independently submitted by a plurality of clients, respectively, verifying whether execution content of the plurality of instances of the blockchain storage request match, and in response to verifying the match, storing the blockchain storage request within a block among a hash-linked chain of blocks.

Abstract: Methods are presented for providing dynamic search filter suggestions that are updated and ranked based on the user filter selections. One method includes detecting a query received in a user interface (UI), calculating, by a search-candidate model, first search results, and calculating, by a suggestions model, first filter suggestions for filter categories to filter responses to the query. The suggestions model is obtained by training a machine-learning algorithm utilizing pairwise learning-to-rank modeling. The first search results and the first filter suggestions are presented in the UI. When a selection in the UI of a filter suggestion is detected, the search-candidate model calculates second search results for the filter categories based on the query and the selected filter suggestion, and the suggestions model calculates second first filter suggestions based on the query and the selected filter suggestion. The second search results and the second filter suggestions are presented in the UI.

Abstract: Systems and methods for determining a query for a data store are described. A natural language text may be analyzed using heuristic processing and one or more machine learning models. Query parameters may be determined from the heuristic processing and machine learning and combined to form a query in a query language. In the heuristic processing, parsing rules may be used to remove conditional terms to generate a base question. The base question may be input to the one or more machine learning model to generate a base query which may be combined with query parameters related to the conditional terms.

Abstract: The present disclosure provides an apparatus for and method of generating a context category dataset. According to some embodiments, the present disclosure provides a context category dataset generating apparatus and method which predict a context category to which a user-inputted hashtag belongs, receive from the user the user's context category to which the hashtag belongs, and generate and update the context category dataset.

Abstract: A method for execution by a query processing system includes determining a query request that indicates a plurality of operators, where the plurality of operators includes at least one relational algebra operator and further includes at least one non-relational operator. A query operator execution flow is generated from the query request that indicates a serialized ordering of the plurality of operators. A query resultant of the query is generated by facilitating execution of the query via a set of nodes of a database system that each perform a plurality of operator executions in accordance with the query operator execution flow, where a subset of the set of nodes each execute at least one operator execution corresponding to the at least one non-relational operator in accordance with the execution of the query.

Abstract: Joining data using a disjunctive operator using a lookup table is described. An example computer-implemented method can include receiving a query with a set of conjunctive predicates and a set of disjunctive predicates. The method may also include generating a lookup table for each predicate in the sets of conjunctive predicates and disjunctive predicates. The method, for each row in a probe-side table, may also further include looking up a value associated with that row in each of the lookup tables and adding the row to a results set when there is a match. Additionally, the method may also include returning the results set.

Abstract: The present application discloses a method, an apparatus, a device, and a storage medium for intention recommendation, which relates to the field of big data, artificial intelligence, intelligent search, information flow and deep learning technologies in the field of computer technologies. A specific implementation scheme includes: receiving an intention query request carrying an intention keyword and a user identification, determining a first recommendation list according to the intention keyword and a pre-configured intention repository, where the intention repository includes at least one tree-shaped intention set, and each tree-shaped intention set includes at least one graded intention, processing intentions in the first recommendation list by using intention strategy information corresponding to the user identification to obtain a target recommendation list and output it.

Abstract: Embodiments of the invention relate to a method for managing subscriptions. The method includes initiating execution of a first subscription, in response to the initiating, obtaining a first subscription barrier associated with a first subscription, making, using the first subscription barrier, a first determination to block execution of a first query request associated with the first subscription, and in response to the first determination, ceasing execution of the first subscription.

Abstract: A method may include responding to a query to retrieve data from a database by identifying a plurality of query execution plans. An overall robustness value may be determined for each query execution plan. The overall robustness value of a query execution plan may correspond to a sum of individual robustness values for each operator included in the query execution plan. Each operator may have an individual robustness value that corresponds to a first change in a total cost of a query execution plan including the operator relative to a second change in an output cardinality of the operator. One of the plurality of query execution plans may be selected based on the overall robustness value of each of the plurality of query execution plans. The query may be executed by performing a sequence of operators included in the selected one of the plurality of query execution plan.

Abstract: Systems and methods are described for exporting bucket data from one or more buckets to one or more worker nodes. The system can identify data from different bucket data from buckets stored in a data intake and query system that is to be processed by one or more worker nodes. The system can allocate one or more execution resources, such as a processing pipeline, to process and export the bucket data from the buckets. The system can assign bucket data corresponding to individual buckets to the execution resource based on a bucket distribution policy. The indexer can export the bucket data to the worker nodes for further processing based on the bucket data-execution resource assignment.

Abstract: Herein are techniques for dynamic aggregation of results of a database request, including concurrent grouping of result items in memory based on quasi-dense keys. Each of many computational threads concurrently performs as follows. A hash code is calculated that represents a particular natural grouping key (NGK) for an aggregate result of a database request. Based on the hash code, the thread detects that a set of distinct NGKs that are already stored in the aggregate result does not contain the particular NGK. A distinct dense grouping key for the particular NGK is statefully generated. The dense grouping key is bound to the particular NGK. Based on said binding, the particular NGK is added to the set of distinct NGKs in the aggregate result.

Abstract: Method and apparatus for stress management in a searchable data service. The searchable data service may provide a searchable index to a backend data store, and an interface to build and query the searchable index, that enables client applications to search for and retrieve locators for stored entities in the backend data store. Embodiments of the searchable data service may implement a distributed stress management mechanism that may provide functionality including, but not limited to, the automated monitoring of critical resources, analysis of resource usage, and decisions on and performance of actions to keep resource usage within comfort zones. In one embodiment, in response to usage of a particular resource being detected as out of the comfort zone on a node, an action may be performed to transfer at least part of the resource usage for the local resource to another node that provides a similar resource.

Abstract: Methods and apparatuses in a cloud-based database management system are described. Data in a database are stored in a plurality of pages in a page store of the database. A plurality of redo log records are received to be applied to the database. The redo log records within a predefined boundary are parsed to determine, for each given redo log record, a corresponding page to which the given log record is to be applied. The redo log records are reordered by corresponding page. The reordered redo log records are stored to be applied to the page store of the database.

Abstract: Distributed pseudo-random subset generation includes obtaining a data-query indicating a first table having a first column including unique values, a second table having a second column including unique values, a join clause joining the first table and the second table on the first column and the second column, and a limit value, pseudo-random filtering the first table to obtain left intermediate data and left filtering criteria, pseudo-random filtering the second table to obtain right intermediate data and right filtering criteria, obtaining intermediate results data by full outer joining the left intermediate data and the right intermediate data, obtaining results data by filtering the intermediate results data using most-restrictive filtering criteria among the left filtering criteria and the right filtering criteria, and outputting the results data, wherein outputting the results data includes limiting the cardinality of rows of the results data to be at most the limit value.

Abstract: Systems and methods for processing natural language inputs to determine user intents using an insights repository are provided. An insights repository system is configured to build an insights repository as a data structure representing a plurality of entities and relationships among those various entities. The insights repository system may receive information from various sources via an event stream, and may process the information using event rules. Based on the application of the event rules, the system may configure an insights repository data structure representing various entities, relationships between various entities, and the strengths of relationships between various entities. After the insights repository is created, consumers may execute queries against the insights repository.

Abstract: Methods and systems for presenting a proactive audiovisual query using a virtual assistant software application. The methods correspond to retrieving user experience data associated with a user profile for the virtual assistant software application, wherein the user experience data define one or more user experience events for the user profile. The methods further include retrieving subject matter domain data associated with the one or more user experience events, wherein the subject matter domain data define one or more probabilistic event effects for the one or more user experience events and determining, based on the user experience data and the subject matter domain data, one or more optimal query items, wherein the one or more optimal query items are associated with at least one of the one or more probabilistic event effects. The methods further include generating the proactive audiovisual query in accordance with the one or more optimal query items.

Abstract: Briefly stated, the invention is directed to retrieving a semantically matched knowledge structure. A question and answer pair is received, wherein the answer is received from a query of a search engine. A question is constraint-matched with the answer based on maximizing a plurality of constraints, wherein at least one of the plurality of the constraints is a similarity score between question and answer, wherein the constraint matching generates a matched sequence. For one or more answer sequences, a subsequence is found that are not parsed as answer slots. Query results are obtained from another search engine based on a combination of the answer or question, and the non-answer subsequence. And a KB based is refined on the query results and the constraint matching and based on a neural network training, for a further subsequent semantic matching, wherein the KB includes a dense semantic vector indication of concepts.

Abstract: Methods and systems for improved categorization of queries are disclosed. In one aspect, first queries having query results limited to a particular category are identified. Second queries including a first query are also identified. For identified second queries, a pattern is generated based on the second query and the included first query, and a rank of the first query's particular category in results for the second query and a percentage of the second query's results having a category equivalent to the included first query's particular category are determined. The ranks and percentages are aggregated for matching patterns, and second patterns with aggregated ranks and percentages meeting a criterion are determined. Results for a third query are limited to a category equivalent to a particular category for a first query included in the third query, and then transmitted over a computer network to a client device.

Abstract: An automated method for search includes: receiving search characters and a viewport location on a displayed geographic area corresponding to an area of interest; simultaneously searching a first entry source and a second entry source to obtain corresponding first suggested entries and second suggested entries, where the first suggested entries correspond to geographic locations that are closer to the viewport location; ranking the first suggested entries according to first rules of relevancy, and generating first ranked suggested entries; ranking the second suggested entries according to second rules of relevancy, and generating second ranked suggested entries; combining the first and second ranked suggested entries into a combined set of suggested entries, and ranking the combined set of suggested entries according to combined rules of relevancy, and generating combined ranked suggested entries; and transmitting the combined ranked suggested entries to a user for selection of a desired type ahead entry.

Abstract: In some embodiments, a method includes determining a position for a search query and a position for each audience record from multiple audience records in an embedding space. The method further includes receiving multiple device records, each associated with an audience record. The method further includes determining multiple keywords, each associated with an audience record and determining a position for each keyword in the embedding space. The method further includes calculating a first distance between the position of the search query in the embedding space and the position of each audience record in the embedding space. The method further includes calculating a second distance between the position of the search query in the embedding space and the position of each keyword in the embedding space. The method further includes ranking each audience record based on the first distance and the second distance.

Abstract: Data exploration as search over automated pre-generated plot objects can include data analytics systems with automated data mining and simplified user experience front ends. A computer-implemented method, that can be performed by the described data analytics systems, includes receiving a request for plots or plot types of a specified criteria; searching a plot object resource for plots relevant to the specified criteria, the plot object resource comprising an indexed repository of available plots; sorting and ranking the plots or plot types according to associated scores, the associated scores for each plot being based on information theoretic metrics relevant to a measure of interest; and providing plots satisfying a criteria of the sorting and the ranking to a source of the request.

Abstract: Systems and methods for automated and explainable machine learning to generate seamlessly actionable insights by generating explainable personas directly from customer relationship management systems are disclosed. The personas are defined as a collection of segments, scored by likelihood to generate good opportunities, accompanied ranked profile attribute importance, with descriptive names and summaries, associated human and database readable queries which have been generated to optimally find cluster candidates in a broader data universe. Such a system would effectively and accurately model the composition of past clients, perform the categorization in an explainable way such that actions can be taken on the information to have predictable results.

Abstract: Examples described herein may perform various operations based on mood congruency. An example method involves accessing, by a processor, from a database, a score that represents a degree of congruency between a first mood vector that describes first media data and a second mood vector that describes second media data, wherein the score is generated based on (i) a first value that the first mood vector associates with a first mood, (ii) a second value that the second mood vector associates with a second mood, and (iii) a degree of congruency between the first and second moods, based on the score, comparing, by the processor, a first characteristic of the first media data, other than the first mood, with a second characteristic of the second media data, other than the second mood, and based at least in part on an output of the comparing, providing an indicator to a module.

Abstract: Disclosed in some examples are methods, systems, and machine readable mediums which utilize volume to improve the ordering of search results for various information retrieval systems. This improves relevance as volume is a proxy for interest. As volume changes over time, the relevance of a particular result to a particular search query will increase or decrease over time.

Abstract: Systems and methods are provided for using point-in-time architecture (PTA) databases. An exemplary method includes: entering first data, received from a first data source, into a first PTA database; receiving a first instruction to process the first data using a first statistical operation; executing the first statistical operation for the first data, resulting in first output data; filtering the first output data based on a user-selected attribute; and performing multiple stages of a data processing operation for the first output data.

Abstract: A method for managing big metadata using columnar techniques includes receiving a query request requesting data blocks from a data table that match query parameters. The data table is associated with system tables that each includes metadata for a corresponding data block of the data table. The method includes generating, based on the query request, a system query to return a subset of rows that correspond to the data blocks that match the query parameters. The method further includes generating, based on the query request and the system query, a final query to return a subset of data blocks from the data table corresponding to the subset of rows. The method also includes determining whether any of the data blocks in the subset of data blocks match the query parameters, and returning the matching data blocks when one or more data blocks match the query parameters.

Abstract: A method and apparatus for mining a competition relationship between POIs. An embodiment of the method includes: acquiring a graphlet mining result obtained by mining map retrieval data of users which encompasses attribute information of retrieved target POIs, the graphlet mining result encompassing occurrence frequencies of respective preset situations, and a preset situation comprising: conforming to attribute information of POIs represented by a corresponding preset graphlet and a preset association relationship between attribute information of at least two POIs; for a first and second POI, determining an occurrence frequency of a preset situation corresponding to a preset graphlet where attribute information of the first and second POI co-occur, and generating a relationship feature of the first and second POI; and inputting the relationship feature into a pre-trained relationship prediction model to obtain a competition relationship prediction result of the first POI and the second POI.

Abstract: A method for accessing information. The information is received by a computer system from sources for distribution to client computer systems. A piece of the information received from the sources without temporal data is identified by the computer system. The temporal data for the piece of the information based on a policy is identified by the computer system. The temporal data is associated with the piece of the information by the computer system, enabling analyzing the information by a client computer system with increased accuracy.

Abstract: Techniques for request throttling in a provider network environment are described. A throttle handler controls whether requests will be processed through maintaining a token-based record, per type of request, having a token value indicative of a number of requests that can be processed over a time period. For a request, the token value of the token-based record corresponding to the request type is updated based on calculating an elapsed time between a last update time of the token-based record and the current time, calculating an intermediate token value as the existing token value plus a value of the elapsed time multiplied by a rate, and updating the token value to be the minimum between the intermediate token value and a burst value. The request is serviced when the updated token value is determined to be greater than or equal to a number of tokens needed to perform the request.

Abstract: A method may include presenting a user interface on a display device of a computing device, the user interface including: a search query input element; a plurality of graph type options; a graph level selection element; and a graph presentation area; receiving a search query inputted into the search query input element, the search query identifying a concept object in an ontology; retrieving data associated with the concept object from a graph database based on selections made in the graph type options and the graph level selection element, the data including a set of result objects related to the concept object; and rendering a hierarchical graph in the graph presentation area, the hierarchical graph illustrating the set of result objects and the concept object as interactive nodes.

Abstract: A DBMS receives a database-access request that includes an instruction to non-destructively read a database table row. The DBMS assigns the request a TSN identifier and creates a TSN image that identifies all TSNs assigned to transactions that are not yet committed. The DBMS traverses a linked list of log entries that identifies a chronological history of transactions performed on the same row. The DBMS infers that the table row currently contains data stored by the most recently logged transaction that is not contained in the TSN image and that has thus been committed. The DBMS then continues to process statements of the transaction based on the assumption that the row contains the inferred value. The DBMS performs this procedure without acquiring a shared lock on the data page or on the index leaf page that points to the table row.

Abstract: A set of unstructured data is analyzed to infer structural elements from the unstructured data, and quantized data quality levels, indicative of data quality in the structural elements, are assigned to the structural elements. A set of structured data is generated to include the structural elements inferred from the unstructured data and associations between respective ones of the structural elements in the set of structured data and the corresponding quantized quality levels assigned to the structural elements. The set of structured data, including the associations between respective ones of the structural elements and the corresponding quantized quality levels assigned to the structural elements, is provided to a user interface application to enable the user interface application to visually display varying data qualities in the set of structured data.

Abstract: A method includes calculating first connection usage information based on a first number of connections being used in a connection pool over a first time interval and adding a first number of database connections to the connection pool based on determining that the first connection usage information satisfies a first usage threshold. Additionally, the method includes calculating second connection usage information based on a second number of connections being used in the connection pool over a second interval. The second interval has been previously reset responsive to adding the first number of database connections to the connection pool. The method also includes determining whether to adjust the number of database connections in the connection pool based on the second connection usage information.

Abstract: Apparatuses and methods are provided to generate customizable databases and/or analyze performance. In an example embodiment, a method of generating customizable databases is provided. The method includes receiving a calculation expression relating to one or more defined characteristics. The calculation expression may be defined by a user. The method also includes loading data into a data warehouse. The data includes at least one of the one or more defined characteristics. The method further includes generating a data cube based on the received calculation expression and the data loaded into the data warehouse. The data cube includes an accessible table. A corresponding apparatus is provided. Additional method and apparatus to analyze performance are also provided.

Abstract: A cloud computing system, which works in cooperation with a plurality of relay devices, is configured to receive measurement data transmitted from each of the relay devices arranged in respective bases and perform accumulation processing in a hierarchical structure of a logical tree form in a measurement database, and perform aggregation analysis processing on the measurement data subjected to the accumulation processing in the respective bases and for each integration target between the bases. The aggregation analysis processing is performed on the measurement data for the each integration target between the bases by recognizing a relationship between the bases under the same starting point on the basis of a measurement unit of a measurement value or a type of a measurement source as for the value in the source accumulated in the database as an ending point of the hierarchical structure.

Abstract: Methods and systems for cross cluster replication are provided. Exemplary methods include: periodically requesting by a follower cluster history from a leader cluster, the history including at least one operation and sequence number pair, the operation having changed data in a primary shard of the leader cluster; receiving history and a first global checkpoint from the leader cluster; when a difference between the first global checkpoint and a second global checkpoint exceeds a user-defined value, concurrently making multiple additional requests for history from the leader cluster; and when a difference between the first global checkpoint and the second global checkpoint is less than a user-defined value, executing the at least one operation, the at least one operation changing data in a primary shard of the follower cluster, such that an index of the follower cluster replicates an index of the leader cluster.

Abstract: Source database precommitted transactions are resolved in a target database of a database replication system when selected source database precommitted transactions are subsequently aborted in the source database.

Abstract: Requests are received for handling by a cloud computing environment which are then executed by the cloud computing environment. While each request is executing, performance metrics associated with the request are monitored. A vector is subsequently generated that encapsulates information associated with the request including the text within the request and the corresponding monitored performance metrics. Each request is then assigned (after it has been executed) to either a normal request cluster or an abnormal request cluster based on which cluster has a nearest mean relative to the corresponding vector. In addition, data can be provided that characterizes requests assigned to the abnormal request cluster. Related apparatus, systems, techniques and articles are also described.

Abstract: The present disclosure provides a method of user clustering, and the method includes: acquiring a clustering condition for a predetermined user group, wherein the clustering condition includes a time selecting condition and an event selecting condition; determining at least one target time period for each user behavior data in a user behavior database based on the time selecting condition; determining association data indicating a relationship between the each user behavior data and each target time period based on the each user behavior data and the each target time period; and selecting target association data for a time period to be monitored based on the time period to be monitored and the event selecting condition, so as to determine a target user belonging to the predetermined user group according to the target association data. The present disclosure also provides an apparatus of user clustering, a computer device and a non-transitory medium.

Abstract: Systems and techniques for indexing and/or querying a database are described herein. Multiple, large disparate data sources may be processed to cleanse and/or combine item data and/or item metadata. Further, attributes may be extracted from the item data sources. The interactive user interfaces allow a user to select one or more attributes and/or other parameters to present visualizations based on the processed data.

Abstract: Systems and methods for automated data curation and presentation are disclosed herein. The system can include a memory including a structured database and a plurality of storage bins. The system can include at least one server that can receive a packetized data file generated from a data file. This packetized data file can include a first packet generated from a content file of the data file, a second packet generated from metadata of the data file, and a third packet generated from a payload of the data file. The at least one server can automatically generate at least one tag for the packetized data file, which at least one tag is automatically generated based at least one key phrase identified in at least the targeted portion of the content file.

Abstract: A computer generates and displays a data visualization in a data visualization user interface according to placement of data fields, from a data source. The data visualization includes visual data marks representing data from the data source. The computer detects a user input to select a visual data mark. In response to detecting the user input, the computer obtains a data model encoding the data source as a tree of logical tables. The computer identifies one or more aggregated data values for the visual data mark, each of the aggregated data values corresponding to a respective data field in the data model. For each of the aggregated data values, the computer retrieves a respective disaggregated set of data rows from a respective logical table containing the respective data field. The computer displays a summary grid, with a respective tab corresponding to each of the retrieved disaggregated sets of data rows.

Abstract: A system and method for visualizing and querying high dimensional data to a user. The system includes a user device, a data-pearls visualization and querying server. The server obtains the high dimensional data from the user device associated with user. The server generates data clusters and sub-divides the data clusters into non-overlapping subsets of data-pearls using a clustering technique. The server selects a shape for each data-pearl by comparing a distance between centroid of a data-pearl and a farthest point from a determined centroid using Lp norm distance measures. The server configures each data-pearl in a three-dimensional plot. The server enables the user to visualize the data-pearls on a screen of the user device. The server queries data based on a query using data dimension technique. The server dimensions data related to the query through determined classifiers based on filtered data after pruning unrelated data to the query.

Abstract: Provided are systems and methods for classifying and tagging records in a record management system using information extracted and analyzed from specific areas or regions of records. A specific area or region of the record may be scanned, and the content disposed therein processed against a plurality of classification templates. Based on proximity to the classification templates, the record may be assigned one or more tags corresponding to the classification templates.

Abstract: A system and associated methodology manages localization data. According to one embodiment of the present invention a primary set of data associated with one or more specific locations is imported and matched to a predefined format. Thereafter external, secondary, data associated with each of the localities listed in the primary data set is collected from a plurality of third party location service providers. With the collected secondary data matched to the same predefined format a comparison is made between the plurality of secondary data sets and the primary data set. Differences between the data sets are identified and the primary data set modified as necessary. Thereafter, normalized data from the modified primary data set is exported to the third party location service providers to enhance consistency and reliable of locational data.

Abstract: Techniques for improved data services are provided. Upon receiving a first request from a first client, a first timeline comprising a first plurality of events for an asset indicated in the first request is generated. Upon receiving a second request from a second client, a second timeline comprising a second plurality of events is generated. A first submission for the asset is provided to the first client, comprising updating a first graphical user interface (GUI) output on a first device of the first client. Upon determining that the first client approved the first submission, a merged timeline is generated based on the first and second timelines, where the merged timeline includes the first and second pluralities of events, comprising: updating the first GUI output on the first device, and a second GUI output on a second device of the second client, to indicate the merged timeline.

Abstract: Systems, devices, and methods of the present invention create a searchable index that includes informative portions of text. In an example, a computer-implemented method creates a discourse tree from a body of text. For each non-terminal node in the discourse tree, the method identifies a rhetorical relationship associated with the non-terminal node. The method labels each terminal node associated with the non-terminal node as either a nucleus or a satellite. The method further accesses a rule associated with the rhetorical relationship, and selects, based on the rule, selects the fragment associated with the nucleus. The method creates a searchable index including the selected fragments.