Abstract: A method includes storing a research paper and a set of candidate resources including media content. The method further includes encoding each of one or more first content fields in the research paper into a first vector based on a first field type associated with each of the one or more first content fields. The method further includes encoding each of one or more second content fields in each of the parsed set of candidate resources into a second vector, based on a second field type associated with each of the one or more second content fields. The method further includes comparing the first vector with the second vector to determine a final set of resources based on the comparison. The method further includes controlling a display screen to output the determined final set of resources and the research paper.

Abstract: A method includes seed expansion using Graph Neural Network (GNN). The method includes retrieving a first plurality of seed accounts based on a first user input indicative of domain-specific information associated with the first plurality of seed accounts. The method further includes determining a second plurality of seed accounts based on the first plurality of seed accounts associated with the domain-specific information, for training a GNN. The method further includes updating the retrieved first plurality of seed accounts and the determined second plurality of seed accounts based on the trained GNN. The method further includes generating a final set of seed accounts associated with the domain-specific information, based on an application of a predefined ranking criteria on the updated first plurality of seed accounts and the second plurality of seed accounts. The method further includes controlling a display screen to output the generated final set of seed accounts.

Abstract: A method includes seed expansion using Graph Neural Network (GNN). The method includes retrieving a first plurality of seed accounts based on a first user input indicative of domain-specific information associated with the first plurality of seed accounts. The method further includes determining a second plurality of seed accounts based on the first plurality of seed accounts associated with the domain-specific information, for training a GNN. The method further includes updating the retrieved first plurality of seed accounts and the determined second plurality of seed accounts based on the trained GNN. The method further includes generating a final set of seed accounts associated with the domain-specific information, based on an application of a predefined ranking criteria on the updated first plurality of seed accounts and the second plurality of seed accounts. The method further includes controlling a display screen to output the generated final set of seed accounts.

Abstract: In an embodiment, operations including extracting a first set of sub-graphs from a first graph. A set of network motifs is extracted from the first graph based on first statistical information associated with the extracted first set of sub-graphs, and a set of network motif instances of the set of network motifs is located. Second statistical information of the set of network motifs is determined based on the set of network motif instances. The second statistical information includes network motif information of a node or edge associated with the set of network motif instances. A graph explanation model is applied on the first graph based on the second statistical information. A first explanation sub-graph for a first class is extracted from the first graph, based on the application of the graph explanation model and a set of regularization constraints. An output representation indicative of the first explanation sub-graph is displayed.

Abstract: Technologies are described to provide autonomous speech and language assessments. In some examples, a method to conduct an autonomous speech and language assessment includes facilitating authoring of an autonomous speech and language assessment. The autonomous speech and language assessment may be interactions between one or more of a conversational module, an analytical module, and a control module, and the interaction may be specified using a visual programming language. The method may also include modeling the autonomous speech and language assessment as a graph using a conversation markup language, and conducting the autonomous speech and language assessment utilizing a conversational virtual agent.

Abstract: In an embodiment, operations include crawling a set of web pages and labeling one or more items of a first web page based on user input. Each item corresponds to a node in a first tree data structure of the first web page. The operations further include generating a first extraction rule to extract a first item from the one or more first items. The first extraction rule includes a first path, in the first tree data structure, for a first node of the first item, and includes first visual information of each node in the first path. The operations further include comparing the first visual information in the first path with second visual information of each of a plurality of candidate nodes in a second tree data structure of a second web page and further refining the first extraction rule to generate a second extraction rule.

Abstract: According to an aspect of an embodiment, operations may include obtaining multiple electronic documents and obtaining a theme text. The method may also include selecting a seed text based on a semantic similarity between the seed text and the theme text. The method may also include changing a seed weight included in a weight vector that is used in identification of topics of the multiple electronic documents. The changed seed weight may bias the identification of topics of the plurality of electronic documents in favor of the seed text as compared to one or more other text strings of the weight vector. The method may also include generating, a representation of a topic model for display to a user, the topic model may be based on the multiple electronic documents and the weight vector.

Abstract: A method of automatically estimating the difficulty level of online content. The method includes receiving, by one or more processors, electronic content which is accessible via an online network and which is at least temporarily stored in at least one non-transitory storage media, extracting, by the one or more processors, structural and non-linguistic information from the electronic content. The method further includes extracting, by the one or more processors, linguistic information from the electronic content. The method further includes generating, by the one or more processors, a difficulty estimation for the content based on the structural and non-linguistic information in addition to the linguistic information. The method further includes generating, by the one or more processors, a recommendation of a subset of the electronic content to a user based on the difficulty estimation.

Abstract: A method of collecting and organizing online resources may include selecting an item from an electronic document. The method may include extracting data related to the item from the electronic document. The method may include identifying whether a first list that includes the item is associated with the electronic document. The method may include refining the first list in response to an identification of the first list. The method may include generating a second list that includes the item in response to a failure to identify the first list. The method may include verifying that the first list is consistent with a user interest. The method may include saving the first list to an electronic data storage device when the first list is consistent with the user interest. The method may include selecting a second item when the first list is inconsistent with the user interest.

Abstract: A method of generating a recommendation for an item is provided. A method may include receiving search results based on a query for an item. The method may also include identifying a plurality of human-curated lists from the search results, wherein each human-curated list of the plurality of human-curated lists includes a plurality of human-recommended items. Further, the method may include determining, via the at least one processor, one or more scores associated with at least one of the plurality of human-curated lists and the plurality of human-recommended items. The method may further include generating, via the at least one processor, a recommendation including at least one recommended item based on the determined one or more scores. Furthermore, the method may include displaying the recommendation for the item.

Abstract: A method of identifying domain-specific accounts is provided. A method may include identifying a plurality of candidate accounts based on a plurality of seed accounts. The method may further include identifying a subset of candidate accounts of the plurality of candidate accounts based on one more account statistics. Further, the method may include clustering each candidate account of the subset of candidate accounts into either a non-domain-specific cluster or a domain-specific cluster based on a plurality of account statistics. The method may also include updating, at least one of the plurality of candidate accounts and the plurality of seed accounts based on at least one of the non-domain-specific cluster and the domain-specific cluster.

Abstract: A general type weight and an individual weight are determined for each of multiple social media accounts. The general type weight may be based on a social media account type corresponding to the social media account. The method may include encoding a mutually reinforcing relationship between the social media accounts and contents promoted by each of the social media accounts. The mutually reinforcing relationship may be encoded as a promotional link between the social media accounts and a content item of the contents promoted by each of the social media accounts. The method may include calculating a basic link strength for each promotional link, where the basic link strength is based on the general type weight and the individual weight for each social media account of multiple social media accounts, and calculating a mutually reinforcing ranking of the multiple social media accounts and contents based on the basic link strengths.

Abstract: An example computer-implemented method may include generating a corpus of content, where the corpus of content comprises content items. The method may also include calculating a ranking score of each content item in the corpus of content items, where the ranking score of a content item is based on one or more of a topic match of the content item, a credit of the content item, and a freshness of the content item. The method may further include aggregating each content item of the content items into one of multiple sections in a reading list tailored for a user, and ranking the content items in each section based on respective ranking scores of the content items. The reading list may then be displayed, for example, for viewing by the user.

Abstract: A method may include obtaining multiple electronic documents and multiple topics associated with the electronic documents. The method may further include determining a similarity between a first topic and a second topic. The first topic may be associated with a first set of electronic documents. The method may further include refining the multiple topics based on the similarity between the first topic and the second topic by associating the first set of the electronic documents with the second topic and removing the first topic from the multiple topics. The method may further include building a document-classifier model by applying machine learning to at least one electronic document associated with each of the refined topics. The method may further include obtaining an electronic document and classifying the electronic document into one of the refined topics using the document-classifier model.

Abstract: A method to identify information about a mathematical formula may include receiving a mathematical formula included in a first electronic document. The method may also include generating a formatted formula from the mathematical formula. The method may also include finding a returned formula that is similar to or the same as the formatted formula in a plurality of mathematical formulae. The method may also include identifying from at least one other electronic document content that describes the returned formula and includes a derivation of the returned formula.

Abstract: A method may include receiving one or more electronic message streams. The method may also include selecting a time range for analysis of the one or more electronic message streams. The method may include analyzing a set of historic data of the one or more electronic message streams. The method may also include identifying one or more bursty phrases in the set of historic data of the one or more electronic message streams in the selected time range. The method may further include clustering the bursty phrases in one or more bursty topics. The method may include generating a rank for each of the one or more bursty topics. The method may also include displaying the one or more bursty topics in a graphical user interface (GUI) in view of the rank. The method may include storing the one or more bursty topics.

Abstract: A method may include obtaining a multiple known objects and a player construct, comparing the player construct to the known objects, and determining whether the comparison exceeds a matching threshold. If a comparison exceeds the matching threshold, the player construct is assigned to a class associated with one of the known objects, and a meaningfulness measurement can be determined. The meaningfulness measurement may be based on a quantity of player constructs, including the first player construct, a similarity between elements in the class, and a similarity between the class and one or more other classes. The player construct may be created using physical blocks embedded with sensors or electronic blocks.

Abstract: A method may include receiving a first vocal input, which may include conversational language describing a portion of a command to be generated for a voice assistant. The method may include determining a structure of the command based on the first vocal input. The method may include generating a template for the command based on the structure. The template may include a particular sequence of segments. The method may include providing a prompt for a second vocal input that includes conversational language. The second vocal input may correspond to at least one segment of the particular sequence. The method may include receiving the second vocal input. The method may include assigning one or more portions of the first and the second vocal input to corresponding segments of the particular sequence. The method may include generating an executable representation of the command, which may include the particular sequence of segments.

Abstract: A method to generate a topic label for a set of electronic documents may include crawling, by a processor, the set of electronic documents. The method may include extracting knowledge points from the set of electronic documents. The method may also include selecting a candidate set of knowledge points from the plurality of knowledge points based on occurrence values. The method may include calculating relatedness scores between each knowledge point in the candidate set of knowledge points. The method may also include calculating hierarchical relationships between each knowledge point in the candidate set. The method may further include calculating comprehensive scores for each knowledge point in the candidate set based on the relatedness scores and the hierarchical relationships. The method may include selecting, from the set of knowledge points, a first candidate knowledge point with the highest comprehensive score as a topic label for the set of electronic documents.

Abstract: A method of generating a recommendation for an item is provided. A method may include receiving search results based on a query for an item. The method may also include identifying a plurality of human-curated lists from the search results, wherein each human-curated list of the plurality of human-curated lists includes a plurality of human-recommended items. Further, the method may include determining, via the at least one processor, one or more scores associated with at least one of the plurality of human-curated lists and the plurality of human-recommended items. The method may further include generating, via the at least one processor, a recommendation including at least one recommended item based on the determined one or more scores. Furthermore, the method may include displaying the recommendation for the item.