Abstract: A system and method for automatically detecting misinformation is disclosed. The misinformation detection system is implemented using a cross-stitch based semi-supervised end-to-end neural attention model which is configured to leverage the large amount of unlabeled data that is available. In one embodiment, the model can at least partially generalize and identify emerging misinformation as it learns from an array of relevant external knowledge. Embodiments of the proposed system rely on heterogeneous information such as a social media post's text content, user details, and activity around the post, as well as external knowledge from the web, to identify whether the content includes misinformation. The results of the model are produced via an attention mechanism.

Abstract: A system and method for automatically detecting misinformation is disclosed. The misinformation detection system is implemented using a cross-stitch based semi-supervised end-to-end neural attention model which is configured to leverage the large amount of unlabeled data that is available. In one embodiment, the model can at least partially generalize and identify emerging misinformation as it learns from an array of relevant external knowledge. Embodiments of the proposed system rely on heterogeneous information such as a social media post's text content, user details, and activity around the post, as well as external knowledge from the web, to identify whether the content includes misinformation. The results of the model are produced via an attention mechanism.

Abstract: Methods, systems, and apparatuses, including computer programs encoded on a computer storage medium, for facilitating analyzing media items and to filter inappropriate media items before distribution to the users. In one aspect, a method includes partitioning digital media items such as videos into segments and/or scenes, and classifying the segments into predetermined classes such as “Violence”, “Conversation”, “Street”, “Nudity”, “Animation”. After classifications have been assigned, the segments are clustered and/or grouped together before presenting the segments belonging to a particular cluster to a rating entity in a single user interface, for further evaluation. After evaluation, the segments of the media items that were approved by the rating entity are used to identify media items for which all the segments were approved by the rating entity before distributing the media items to the users.

Abstract: A device may receive, from a user device, a request for an automation service for a project. The device may analyze the request to identify one or more input keywords. The device may obtain, by using the one or more input keywords to search one or more data sources, project information for a set of projects and/or automation services information for a set of automation services. The device may determine one or more similar projects and/or one or more similar automation services by analyzing the project information and/or the automation services information. The device may provide, to the user device, a recommendation that identifies one or more automation services associated with the one or more similar projects and/or the one or more similar automation services.

Abstract: Methods, systems, and apparatuses, including computer programs encoded on a computer storage medium, for facilitating analyzing media items and to filter inappropriate media items before distribution to the users. In one aspect, a method includes partitioning digital media items such as videos into segments and/or scenes, and classifying the segments into predetermined classes such as “Violence”, “Conversation”, “Street”, “Nudity”, “Animation”. After classifications have been assigned, the segments are clustered and/or grouped together before presenting the segments belonging to a particular cluster to a rating entity in a single user interface, for further evaluation. After evaluation, the segments of the media items that were approved by the rating entity are used to identify media items for which all the segments were approved by the rating entity before distributing the media items to the users.

Abstract: A device may receive a process model and a reference process model. The process model may include a first plurality of nodes corresponding to a first plurality of stages of a particular process. The reference process model may include a second plurality of nodes corresponding to a second plurality of stages of a reference process. The device may map the process model to the reference process model, identify that a first stage corresponds to a second stage, determine a difference between a value, of a metric, associated with the first stage and a value, of the metric, associated with the second stage, cause an interactive user interface to display a visual representation of the difference, identify automation service(s) that compensate for the difference, determine a score for the process model, and cause the interactive user interface to display the score to enable selection of optimal automation service(s) to deploy.

Abstract: A device receives a corpus of text documents, and utilizes feature extraction on a text document, of the corpus of text documents, to generate features from the text document, where the features include binary features, numeric features, and categorical features. The device performs feature engineering on one or more of the binary features, the numeric features, or the categorical features, to generate converted features, and performs feature encoding on the text document, based on the converted features, to represent the text document as a vector with a similarity score for a domain. The device provides the vector with the similarity score for the domain, as training data, to a machine learning model to generate a trained machine learning model, and performs an action using the trained machine learning model.

Abstract: According to an example, intent and bot based query guidance may include receiving a query associated with a domain, and identifying, based on an analysis of the query, an intent of the query by extracting an action associated with the query and an entity associated with the query. An intent model associated with the query may be generated based on a mapping of the action and the entity with a domain model of the domain. An intent domain specific language representation of the intent model associated with the query may be generated. Based on an analysis of the intent domain specific language representation, a plurality of bots may be identified, and a parameterized bot chain may be generated to respond to the query. A runtime binding of bots of the parameterized bot chain may be performed, and invoked to generate the response to the query.

Abstract: A device receives a corpus of text documents, and utilizes feature extraction on a text document, of the corpus of text documents, to generate features from the text document, where the features include binary features, numeric features, and categorical features. The device performs feature engineering on one or more of the binary features, the numeric features, or the categorical features, to generate converted features, and performs feature encoding on the text document, based on the converted features, to represent the text document as a vector with a similarity score for a domain. The device provides the vector with the similarity score for the domain, as training data, to a machine learning model to generate a trained machine learning model, and performs an action using the trained machine learning model.

Abstract: A device may receive a process model and a reference process model. The process model may include a first plurality of nodes corresponding to a first plurality of stages of a particular process. The reference process model may include a second plurality of nodes corresponding to a second plurality of stages of a reference process. The device may map the process model to the reference process model, identify that a first stage corresponds to a second stage, determine a difference between a value, of a metric, associated with the first stage and a value, of the metric, associated with the second stage, cause an interactive user interface to display a visual representation of the difference, identify automation service(s) that compensate for the difference, determine a score for the process model, and cause the interactive user interface to display the score to enable selection of optimal automation service(s) to deploy.

Abstract: A method may include receiving, by a device, an input sample of textual content. The method may include identifying, by the device, a comparison sample that is semantically similar to the input sample. The comparison sample may be identified based on a similarity score, of the comparison sample and the input sample, satisfying a semantic similarity threshold. The method may include identifying, by the device, a plurality of output samples of textual content based on acceptance information corresponding to the plurality of output samples and the comparison sample. The acceptance information may be determined based on a user input regarding similarity or relevance of the plurality of output samples and the comparison sample, and the user input may be received before the input sample is received.

Abstract: A method may include receiving a plurality of samples that include textual content. The method may include extracting unit values, corresponding to structural units, from the plurality of samples. The structural units may identify characteristics of the plurality of samples to be used to identify pattern information. The pattern information may identify unit values that are shared between at least two samples of the plurality of samples. The method may include generating one or more structural representations based on the unit values. The one or more structural representations may identify the pattern information. The method may include generating one or more additional samples based on the one or more structural representations. The one or more additional samples may include at least one of the unit values, and may be generated based on the pattern information. The method may include outputting the one or more additional samples.

Abstract: A device may receive a set of first samples of textual content. A device may identify a set of clusters of first samples of the set of first samples. A device may identify a pattern of occurrence based on the set of clusters. The pattern of occurrence to identify two or more clusters, of the set of clusters, based on an order in which first samples associated with the two or more clusters were generated or received. A device may receive one or more second samples of textual content. A device may determine that the one or more second samples are semantically similar to one or more corresponding clusters associated with the pattern of occurrence. A device may identify a predicted sample based on the pattern of occurrence and the one or more corresponding clusters. A device may perform an action based on identifying the predicted sample.

Abstract: A device may receive, from a user device, a request for an automation service for a project. The device may analyze the request to identify one or more input keywords. The device may obtain, by using the one or more input keywords to search one or more data sources, project information for a set of projects and/or automation services information for a set of automation services. The device may determine one or more similar projects and/or one or more similar automation services by analyzing the project information and/or the automation services information. The device may provide, to the user device, a recommendation that identifies one or more automation services associated with the one or more similar projects and/or the one or more similar automation services.

Abstract: A device may obtain text to be processed to extract constraints corresponding to an object in the text. The constraints may define values permitted to be associated with the object. The device may extract the constraints based on identifying patterns in the text. The device may generate, based on the constraints, positive test data and negative test data for testing values for the object. The positive test data may include a first value that satisfies each of the constraints, and the negative test data may include a second value that violates at least one of the constraints. The device may provide information that identifies the positive test data and the negative test data.

Abstract: According to an example, intent and bot based query guidance may include receiving a query associated with a domain, and identifying, based on an analysis of the query, an intent of the query by extracting an action associated with the query and an entity associated with the query. An intent model associated with the query may be generated based on a mapping of the action and the entity with a domain model of the domain. An intent domain specific language representation of the intent model associated with the query may be generated. Based on an analysis of the intent domain specific language representation, a plurality of bots may be identified, and a parameterized bot chain may be generated to respond to the query. A runtime binding of bots of the parameterized bot chain may be performed, and invoked to generate the response to the query.

Abstract: A device may receive a set of first samples of textual content. A device may identify a set of clusters of first samples of the set of first samples. A device may identify a pattern of occurrence based on the set of clusters. The pattern of occurrence to identify two or more clusters, of the set of clusters, based on an order in which first samples associated with the two or more clusters were generated or received. A device may receive one or more second samples of textual content. A device may determine that the one or more second samples are semantically similar to one or more corresponding clusters associated with the pattern of occurrence. A device may identify a predicted sample based on the pattern of occurrence and the one or more corresponding clusters. A device may perform an action based on identifying the predicted sample.

Abstract: A method may include receiving, by a device, an input sample of textual content. The method may include identifying, by the device, a comparison sample that is semantically similar to the input sample. The comparison sample may be identified based on a similarity score, of the comparison sample and the input sample, satisfying a semantic similarity threshold. The method may include identifying, by the device, a plurality of output samples of textual content based on acceptance information corresponding to the plurality of output samples and the comparison sample. The acceptance information may be determined based on a user input regarding similarity or relevance of the plurality of output samples and the comparison sample, and the user input may be received before the input sample is received.

Abstract: Functional use-case generation may include determining whether a requirements context is available. In response to a determination that the requirements context is available, the requirements context may be determined as a task context and as a rule context for a requirements sentence of a requirements document. The task context and the rule context may be used to select a functional model from a plurality of functional models. A functional use-case that includes an entity that is to perform a task based on a rule may be generated. Further, in response to a determination that the requirements context is not available, a functional model may be selected from the plurality of functional models based on process context, where the functional model includes a process related to the process context, and the functional model that includes the process related to the process context may be used to generate the functional use-case.

Abstract: A method may include receiving a plurality of samples that include textual content. The method may include extracting unit values, corresponding to structural units, from the plurality of samples. The structural units may identify characteristics of the plurality of samples to be used to identify pattern information. The pattern information may identify unit values that are shared between at least two samples of the plurality of samples. The method may include generating one or more structural representations based on the unit values. The one or more structural representations may identify the pattern information. The method may include generating one or more additional samples based on the one or more structural representations. The one or more additional samples may include at least one of the unit values, and may be generated based on the pattern information. The method may include outputting the one or more additional samples.