Abstract: Using a trained image classification model, a first image captured using an image capture device is classified, the trained image classification model executing in the same system as the image capture device, the classifying producing an inappropriate content classification of the first image. Using a trained feature extraction model, a pattern within a portion of the first image is recognized from the first image, the trained feature extraction model executing in the same system as the image capture device, the pattern predefined as inappropriate content. Based on the classification and the pattern, the first image is adjusted.

Abstract: A computer-implemented method includes extracting, by one or more processors of one or more computing devices, a product family name from each of a plurality of unstructured product titles associated with a plurality of products. The method further includes determining, by the one or more processors, a degree of similarity between model numbers of the plurality of products. The method further includes determining, by the one or more processors, that at least two of the plurality of products are variants of one another by determining that the at least two of the plurality of products have a same extracted product family name and determining that the degree of similarity between the model numbers of the plurality of products is above a predetermined threshold.

Abstract: Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing natural language processing operations using an attention-based text encoder machine learning model that is trained using a multi-task training routine that is associated with two or more training tasks (e.g., a multi-task training routine that is associated with two or more sequential training tasks, a multi-training routine that is associated with two or more concurrent training tasks, and/or the like).

Abstract: Multiple sets of documents for different domains may be used to train multiple domain-specific models. A graph model may be generated to include nodes representing concepts included within the domain-specific models. A white space not including any nodes within the graph model may be identified. Analysis of the white space may be performed based on two or more nodes at periphery of the white space. Words/documents that cover the white space may be generated. Novelty of concepts may be readily assessed using the graph model/white space.

Abstract: A reconfigurable automatic document-classification system and method provides classification metrics to a user and enables the user to reconfigure the classification model. The user can refine the classification model by adding or removing exemplars, creating, editing or deleting rules, or performing other such adjustments to the classification model. This technology enhances the overall transparency and defensibility of the auto-classification process.

Abstract: Methods, systems, and apparatuses, among other things, may provide for an interactive dynamic mapping engine (iDME) for business intelligence, which may interactively obtain information for user devices from sources and schema unknown to the user devices.

Abstract: A computer-implemented method, a computer program product, and a computer system for ranking explanatory variables in multivariate analysis. A computer system extracts words from documents related to categories, creates a histogram of the words in each category, and selects top words in each histogram, where the top words are used as representing words in each category. A computer system generates respective feature vectors of explanatory variable candidates and a feature vector of an objective variable, where a feature vector of a corresponding variable includes elements corresponding to respective ones of the categories and a value of element indicates whether a name of the corresponding variable is included in the top words. A computer system calculates cosine similarity between each of the respective feature vectors of the explanatory variable candidates and the feature vector of the objective variable. A computer system ranks the explanatory variable candidates, based on the cosine similarity.

Abstract: Disclosed are systems and methods that that automatically classify, filter, and reduce large volumes of hosted content data using artificial intelligence technology. The aggregated hosted content data is reduced by representing the hosted content data as sets of data polarity identifiers or data polarity values that correspond to one or more sequencing identifiers that are displayed on a graphical user interface. Hosted content data packets are segmented by labeling the hosted content data packets with a sequencing identifier. The hosted content data packets are processed utilizing neural network technology to classify the hosted content data according to a polarity identifier, polarity value, sentiment identifier, or one or more subject identifiers.

Abstract: Disclosed embodiments may include a system for generating custom content. The system may generate a first webpage comprising first feature(s). The system may transmit the first webpage for display via a graphical user interface (GUI) of a user device. The system may iteratively, until a predetermined threshold is achieved: receive data corresponding to a first user; generate, by an encoder, encoded data based on the data; generate, by a decoder, second feature(s) based on the encoded data; modify the first webpage to generate a second webpage comprising the second feature(s); transmit the second webpage for display via the GUI of the user device; receive user feedback associated with the second webpage; and determine whether the user feedback exceeds the predetermined threshold. Responsive to the predetermined threshold being achieved, the system may transmit the latest version of the second webpage without requesting additional user feedback.

Abstract: A system for enterprise collaboration is associated with an overlay network, such as a content delivery network (CDN). The overlay network comprises machines capable of ingress, forwarding and broadcasting traffic, together with a mapping infrastructure. The system comprises a front-end application, a back-end application, and set of one or more APIs through which the front-end application interacts with the back-end application. The front-end application is a web or mobile application component that provides one or more collaboration functions. The back-end application comprises a signaling component that maintains state information about each participant in a collaboration, a connectivity component that manages connections routed through the overlay network, and a multiplexing component that manages a multi-peer collaboration session to enable an end user peer to access other peers' media streams through the overlay network rather than directly from another peer.

Abstract: A semantic understanding method and apparatus, and a device and a storage medium are provided. The method includes: acquiring a recognition character string that matches speech information; acquiring, from an entity vocabulary library, at least one entity vocabulary respectively corresponding to each recognition character in the recognition character string; and according to a situation of each entity vocabulary hitting the recognition character string, determining a matching entity vocabulary as a semantic understanding result of the speech information. By means of the method, insofar as a completely matching entity vocabulary is not acquired, a matching entity vocabulary can still be determined according to an entity vocabulary library, and semantic information of speech is thus accurately understood; and the method also has relatively high fault tolerance for situations such as wrong words, added words, and omitted words, such that the semantic understanding accuracy of speech information is improved.

Abstract: According to one embodiment, an apparatus includes a first acquisition unit, a second acquisition unit, an identification unit, and an output unit. The first acquisition unit acquires a query image and a query text relating to a target object. The second acquisition unit acquires candidate images of the target object. The identification unit identifies from the candidate images a positive image containing a region demonstrating a similarity to the query image higher than or equal to a first threshold value, and identifies a position of the region in the positive image. The output unit outputs training data including the positive image, information representing the position of the region, and a correct label.

Abstract: In various embodiments, methods and systems for presenting a search interface with search query history based functionality is provided. A search query history store comprising search queries is accessed. The search query history store includes search queries executed in a search computing system. A search query comprises one or more commands. A plurality of search queries retrieved from the search query history store is displayed on the search interface using a placement style. A placement style, such as an indent style, provides a structure for separating and arranging commands of a plurality of search queries displayed. The search interface further provides for receiving a selection of at least a portion of a search query from the plurality of search queries to initiate actions or execute actions based on the selection. The search interface includes a search input interface, such as a search bar, where the selection of the portion of the search query is displayed based on a selected action.

Abstract: Various embodiments relate generally to data science and data analysis, computer software and systems, and control systems to provide a platform to facilitate implementation of an interface, and, more specifically, to a computing and data storage platform that implements specialized logic to predict effectiveness of content in electronic messages as a function, for example, modifiable portions of the content. In some examples, a method may include receiving data signals to cause formation of an electronic message, determining a component of the electronic message, identifying one or more message performance criteria, characterizing the component to identify a component attribute, predicting the component attribute matches at least one of the message performance criteria, and transmitting the electronic message.

Abstract: According to certain embodiments, the present disclosure includes a method for generating a knowledge graph of clinical information for use as a reference model to semantically represent relevant information from clinical encounters. In certain embodiments, the method includes representing a clinical concern as a module of the knowledge graph. In some examples, the method includes associating at least one section node with the module, wherein each section node of the at least one section node corresponds to a clinical concept relevant to the clinical concern and associating at least one topic node with the at least one section node, wherein each topic node of the at least one topic node corresponds to a clinical topic relevant to the clinical concept. And, in certain embodiments the method includes outputting the knowledge graph of clinical information to semantically represent the relevant information from a clinical encounter associated with the clinical concern.

Abstract: Described is a system and method for enabling dynamic selection of a search input. For example, rather than having a static search input box, the search input may be dynamically positioned such that it encompasses a portion of displayed information. An image segment that includes a representation of the encompassed portion of the displayed information is generated and processed to determine an object represented in the portion of the displayed information. Additional images with visually similar representations of objects are then determined and presented to the user.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining summary content for resources in a domain. In one aspect, a method includes accessing a first resource belonging to a particular domain, selecting an anchor in the first resource linking to a second resource belonging to the particular domain, identifying particular text content in the first resource that is subordinate to the anchor that the second resource includes the particular text content that is subordinate to the anchor, based on determining that the second resource includes the particular text content that is subordinate to the anchor, generating a domain template for the particular domain, the domain template specifying a location of the particular text content in the second resource, and determining, for each respective resource belonging to the particular domain having a structure matching the domain template, respective text content for the respective resource.

Abstract: Provided is a method for recommending emoticons, the method including: displaying, by a user terminal, a first emoticon related to first meta information; and displaying, by the user terminal, first and second recommendation groups, the first and second recommendation groups each comprising at least one recommended emoticon corresponding to the first emoticon, wherein at least one recommended emoticon included in the first recommendation group is related to at least one piece of first associated meta information corresponding to the first meta information, wherein at least one recommended emoticon included in the second recommendation group is related to at least one piece of second associated meta information that corresponds to the first meta information and is different from the first associated meta information, and wherein the first meta information and the first and second associated meta information are different.

Abstract: The present disclosure relates to a method for enabling a processing of a dataset of records having a set of attributes. The method comprises: selecting a first attribute of the set of attributes and a subset of one or more second attributes of the set of attributes. Distinct values of the subset of second attributes may be determined from the dataset. For each distinct value of the determined distinct values records of the dataset that have said each distinct value may be identified, and a group of words may be formed from values of the first attribute of the identified records. Distinct word sequences may be identified in the formed groups and a level of presence of each word sequence of the word sequences in each of the formed groups may be determined. At least part of the levels of presence may be provided as metadata.

Abstract: A method of notifying a user of a cloud-based content management platform of a comment made in a document associated with the user includes determining a set of comments associated with documents to which the user has access via the cloud-based content management platform, the set of comments including one or more comments added by other users to each document within a last predetermined time period, selecting one or more comments from the set of comments for notification to the user, the selecting being based at least on interactions of the user with a comment thread associated with each comment from the set of comments and characteristics of the comment thread, wherein a comment thread is a group of a plurality of comments comprising a first comment and a second comment added in reply to the first comment, determining one or more documents associated with the one or more selected comments, and providing a graphical user interface (GUI) of a cloud storage of the user hosted by the cloud-based content managem

Abstract: Embodiments of the disclosure render a context menu upon receiving a context menu indication from a user. The context menu comprises a combination of local context menu items and remote context menu items that are selected based on the context received with the context menu indication. The context menu is created by identifying a set of local and remote context menu item providers and receiving both a rank and HTML based on the received context. The set of local and remote context menu items are ranked according to their respective rank and a number of ranked context menu items are selected. The HTML for the selected context menu items are combined and the combined HTML is rendered to yield the context menu. The context menu is presented to the user.

Abstract: Methods, apparatus, and processor-readable storage media for implementing a message management framework using machine learning techniques are provided herein. An example computer-implemented method includes processing a message comprising identifying at least one message type of the message; determining, based at least in part on the at least one identified message type, one or more message failures by applying one or more machine learning-based rules to at least a portion of the message; determining one or more remedial actions by processing, using one or more machine learning techniques, the at least one identified message type, the one or more determined message failures, and multiple attributes of the message; and performing one or more automated actions based at least in part on the one or more determined remedial actions.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, to detect system problems using social media. One of the method includes monitoring posts on a plurality of social network sites. The method includes identifying posts that include at least one of a plurality of keywords, at least some of the keywords indicative of system problems. The method includes tagging the identified posts. The method includes determining, based on the tagged posts, that a system problem is occurring. The method also includes raising an alert regarding the system problem.

Abstract: An information processing apparatus includes an acquisition unit that acquires a text being input, and an output unit that, when search information for searching for target information is created or updated, from suggestion information created or updated by being extracted from the target information or the search information, extracts the suggestion information corresponding to the text acquired by the acquisition unit, and outputs display information for displaying the extracted suggestion information with a suggestion content predetermined for each operation authority of a user.

Abstract: The present disclosure can provide an information processing apparatus capable of increasing the amount of information that can be displayed. The information processing apparatus according to an aspect of the present disclosure includes: a morpheme analysis unit that analyzes a morpheme group included in a document; an importance degree calculation unit that calculates a degree of importance of the morpheme group; and a display period determination unit that determines, based on the degree of importance of the morpheme group, a display period of the morpheme group when the morpheme group is displayed on a display unit.

Abstract: A method modifies a social media post in response to determining that a hashtag for the social media post diverges from how the hashtag is used by other social media posts. A computing device receives a social media post, which contains a hashtag, and analyzes the social media post using natural language processing and a sentiment analysis to determine a sense and sentiment of the social media post. The computing device receives a plurality of reference social media posts, where each of the reference social media posts contains the hashtag, and utilizes the natural language processing and the sentiment analysis to determine a sense and sentiment of the reference social media posts. In response to determining that the sense and sentiment of the social media post do not match the sense and sentiment of the reference social media posts, the computing device causes the social media post to be altered.

Abstract: A computer implemented method, computerised system and computer readable recording medium are disclosed.

Abstract: Systems and methods for receiving a search query and suggesting one or more alternative queries based on the received search query when processing an electronic search are disclosed. The method includes receiving, by a server, an electronic search query from a user device of a user; obtaining one or more search results responsive to the electronic search query; obtaining a suggested electronic search query based on the electronic search query as received; the suggested electronic search query determined based on information representative of an importance of the electronic search query; generating an electronic response including the one or more search results and the suggested electronic search query for display on a display of the user device; and providing the electronic response including the one or more search results and the suggested electronic search query as a hyperlink to the user device for display on the display of the user device.

Abstract: In one embodiment, the invention can be a method for assessing unstructured comments, the method including providing reference data generated from a first set of unstructured comments from a first group; receiving a second set of unstructured comments from a second group; identifying a significant word within each unstructured comment of the second set of unstructured comments; for each significant word identified within the second set of unstructured comments, determining a frequency of occurrence of the significant word; and generating a visualization including a portion of the identified significant words, wherein for each visualized significant word, a first aspect of an appearance of the significant word is based on an extent to which the frequency of occurrence deviates from a frequency of occurrence of the significant word in the first set of unstructured comments.

Abstract: Techniques for identifying relevant, natural language documents within a data repository. An electronic document specifying natural language text describing an issue with a complex system is received. One or more portions are removed from the electronic document, where the portions are determined to satisfy one or more predefined filtering rules. A first set of terms are determined using a data dictionary structure, and a second set of terms are determined, where at least one term in the second set of terms satisfies at least one predefined pattern matching rule. A third set of terms are determined by processing the electronic document as an input to a machine learning model trained to recognize relevant terms within the electronic document. A synthesized electronic document is generated from the first, second and third sets of terms, and the synthesized electronic document is used to identify a set of relevant documents within the data repository.

Abstract: Various embodiments relate generally to data science and data analysis, computer software and systems, and control systems to provide a platform to facilitate implementation of an interface, and, more specifically, to a computing and data storage platform that implements specialized logic to predict effectiveness of content in electronic messages as a function, for example, modifiable portions of the content. In some examples, a method may include receiving data signals to cause formation of an electronic message, determining a component of the electronic message, identifying one or more message performance criteria, characterizing the component to identify a component attribute, predicting the component attribute matches at least one of the message performance criteria, and transmitting the electronic message.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, to detect system problems using social media. One of the method includes monitoring posts on a plurality of social network sites. The method includes identifying posts that include at least one of a plurality of keywords, at least some of the keywords indicative of system problems. The method includes tagging the identified posts. The method includes determining, based on the tagged posts, that a system problem is occurring. The method also includes raising an alert regarding the system problem.

Abstract: Systems and methods are provided for generating training data from queries and user interactions associated with media collections related to the queries, and training a machine learning model using the generated training data to generate a trained machine learning model. The systems and methods further provide for receiving a prediction request comprising a query for relevant media collections, analyzing the query to determine query features, determining a plurality of media collections for the query, analyzing the plurality of media collections to determine media collection features for each media collection of the plurality of media collections, and generating, using the trained machine learning model, a semantic matching score for each media collection of the plurality of media collections based on matching the query features to the media collection features for each media collection of the plurality of media collections.

Abstract: A context-based operation determination apparatus includes a context information obtainer configured to obtain context information; a context awareness database constructed using data related to a previously performed action; a processor configured to determine a recommendation index that corresponds to a recommended operation and the recommended operation using the context information, and data obtained from the context awareness database; and a selection history database constructed using the final operation, the recommendation index, and an actual operation in a state where the final operation is presented, where the processor verifies the recommended operation using the selection history database.

Abstract: This disclosure describes methods and systems for searching documents in a multi-tenant hosting environment. According to embodiments, to conserve hardware resources, a plurality of documents associated with a plurality of tenants may be mapped to the same search index in the multi-tenant hosting environment. In order to search documents associated only with a single tenant in the multi-tenant hosting environment, a tenant identifier is prepended to every key stored in the search index that is associated with the plurality of documents of the single tenant. Moreover, where one document links to another document within the multi-tenant hosting environment, the link is stored in a web graph when a source tenant identifier matches a target tenant identifier for the link. According to embodiments, when conducting a search, the link is resolved only if the link is stored in the web graph.

Abstract: A computer-implemented method, computerized apparatus and computer program product for minimum coordination passage scoring. Given a candidate passage in a document collection potentially matching a query received, a set of overlapping terms between the candidate passage and the query is determined. For each overlapping term in the set, a first measure of a weight of the term in the query, a second measure of a weight of the term in the candidate passage, and a third measure of a specificity of the term in the document collection are calculated. A function of the first and second measure is evaluated to obtain a value reflecting a condition on the relation therebetween. A minimum coordination score representing a relative similarity between the candidate passage and the query is determined based on the value and the first, second and third measures obtained for each of the overlapping terms.

Abstract: Embodiments of the present invention disclose a method, a computer program product, and a computer system for extracting natural language relations between entities. A computer receives a configuration for associating one or more natural language questions with one or more entities and identifies the one or more entities annotated within a document. The computer answers the natural language questions associated with the identified one or more entities based on context surrounding the identified one or more entities. The computer may further transmit the natural language questions associated with the identified one or more entities and the surrounding context to a question and answer service, then receive answers to the natural language questions from the question and answer service. The computer may further determine whether the received answers correctly describe the relation between the identified one or more entities and other entities within the extracted surrounding context.

Abstract: Online advertising campaigns are operated responsive to keyword trends. The keywords used by a group of browsers is analyzed periodically over time. A list of the most frequently used keywords is separated into those that have previously appeared on the list, referred to herein as the stable keywords, and those that are newly emerging, referred to herein as the trending keywords. The advertiser selects at least one advertising creative that the advertiser associates with the stable keywords, referred to herein as the stable creative, and at least one advertising creative that the advertiser associates with the trending keywords, referred to herein as the trendy creative. The advertising system then operates the online advertising campaign to deliver the respective stable and trendy creatives in proportion to the frequency of use of the trending versus the stable keywords.

Abstract: A search engine optimization system is provided with an on-line social network system. The on-line social network system includes or is in communication with a search engine optimization (SEO) system that is configured to prioritize keywords (potential search terms) based on their respective predicted contribution to the ranking of JSERPs. The value of a job-related keyword is expressed as a priority score assigned to that keyword. The SEO system generates priority scores for different keywords, using a probabilistic model that takes into account a value expressing how likely the keyword is to be included in a search query as a search term and/or a value expressing how likely is a search that includes the keyword as a search term is to produce relevant results.

Abstract: In some embodiments, apparatus, systems and methods are disclosed for interfacing with systems and networks, and the data stored thereon, including navigation of same.

Abstract: A search engine optimization system is provided with an on-line social network system. The on-line social network system includes or is in communication with a search engine optimization (SEO) system that is configured to prioritize keywords (potential search terms) based on their respective predicted contribution to the ranking of JSERPs. The value of a job-related keyword is expressed as a priority score assigned to that keyword. The SEO system generates priority scores for different keywords, using a probabilistic model that takes into account a value expressing how likely the keyword is to be included in a search query as a search term and/or a value expressing how likely is a search that includes the keyword as a search term is to produce relevant results.

Abstract: A search engine optimization system is provided with an on-line social network system. The on-line social network system includes or is in communication with a search engine optimization (SEO) system that is configured to prioritize keywords (potential search terms) based on their respective predicted contribution to the ranking of JSERPs. The value of a job-related keyword is expressed as a priority score assigned to that keyword. The SEO system generates priority scores for different keywords, using a probabilistic model that takes into account a value expressing how likely the keyword is to be included in a search query as a search term and/or a value expressing how likely is a search that includes the keyword as a search term is to produce relevant results.

Abstract: In various embodiments, methods and systems for presenting a search interface with search query history based functionality is provided. A search query history store comprising search queries is accessed. The search query history store includes search queries executed in a search computing system. A search query comprises one or more commands. A plurality of search queries retrieved from the search query history store is displayed on the search interface using a placement style. A placement style, such as an indent style, provides a structure for separating and arranging commands of a plurality of search queries displayed. The search interface further provides for receiving a selection of at least a portion of a search query from the plurality of search queries to initiate actions or execute actions based on the selection. The search interface includes a search input interface, such as a search bar, where the selection of the portion of the search query is displayed based on a selected action.

Abstract: In one embodiment, the invention can be a method for assessing unstructured comments, the method including providing reference data generated from a first set of unstructured comments from a first group; receiving a second set of unstructured comments from a second group; identifying a significant word within each unstructured comment of the second set of unstructured comments; for each significant word identified within the second set of unstructured comments, determining a frequency of occurrence of the significant word; and generating a visualization including a portion of the identified significant words, wherein for each visualized significant word, a first aspect of an appearance of the significant word is based on an extent to which the frequency of occurrence deviates from a frequency of occurrence of the significant word in the first set of unstructured comments.

Abstract: Methods of encoding voice data for loading into an artificial intelligence (AI) integrated circuit are provided. The AI integrated circuit may have an embedded cellular neural network for implementing AI tasks based on the loaded voice data. An encoding method may generate a two-dimensional (2D) frequency-time array from an audio waveform, use the 2D frequency-time array to generate a set of 2D arrays to approximate the 2D frequency-time array, load the set of 2D arrays into the AI integrated circuit, execute programming instructions contained in the AI integrated circuit to feed the set of 2D arrays into the embedded cellular neural network in the AI integrated circuit to generate a voice recognition result, and output the voice recognition result. The encoding method also trains a convolution neural network (CNN) and loads the weights of the CNN into the AI integrated circuit for implementing the AI tasks.

Abstract: A system and method for fast concept search in multiple documents where the concept is expressed by plurality of words, all of which have to be in the same sentence and within specified range. The system automatically finds equivalent expressions of the same concept, and returns as search results all documents in which the concept is contained.

Abstract: Methods and arrangements for managing development of information extraction rules. One or more documents are opened for extraction. An interface is provided to create a label and thereupon label a portion of the document. The created label is stored, and an extractor is developed based on the labeling. A test interface is provided for the extractor, and results of a test conducted through the test interface are displayed. The extractor is exported. In accordance with at least one embodiment, developers are presented with eased automated guidance to write extractors, which thereby reduces an overall manual effort involved in extractor development. Generally, a focused, tutorial-type environment serves as a guide based on previously developed best practices.

Abstract: A system, method, and computer-readable medium are disclosed for identifying paraphrases in a natural language processing (NLP) system comprising: receiving a first phrase and a second phrase by a system; analyzing the first phrase and the second phrase to provide a semantic and structural hierarchical comparison assessment, the semantic and structural hierarchical comparison assessment having an associated semantic and structural hierarchical comparison assessment value; and determining whether the semantic and structural hierarchical comparison assessment value exceeds a predetermined paraphrase equivalency criteria; and, responsive to determining the semantic and structural hierarchical comparison assessment value exceeds the predetermined paraphrase equivalency criteria, classifying the second phrase as being a rewording of the first phrase.

Abstract: In an example, a granularity of title similarity is determined, wherein the granularity of title similarity is a level at which social networking data should be filtered to identify titles similar to a target title. Then a weighted graph of titles is constructed at the granularity of title similarity, wherein each node in the weighted graph is a title and a directed edge exists in the weighted graph between a first node and a second node if the social networking data, at the granularity of title similarity, indicates that a transition occurred wherein a member who held a position with a title corresponding to the first node transitioned to a position with a title corresponding to the second node, wherein each directed edge contains a weight indicating a strength of relationship between nodes. The weighted graph of locations is traversed from a node corresponding to the target title in order to identify titles similar to the target title.

Abstract: In an example, a granularity of location similarity is determined, wherein the granularity of location similarity is a level at which social networking data should be filtered to identify locations similar to a target location. Then a weighted graph of locations at the granularity of location similarity is constructed, wherein each node in the weighted graph is a location and a directed edge exists in the weighted graph between a first node and a second node if the social networking data, at the granularity of location similarity, indicates that a transition occurred wherein a member who resided at a location corresponding to the first node transitioned to reside to a location corresponding to the second node. The weighted graph of locations is traversed from a node corresponding to the target location in order to identify locations similar to the target location.