Abstract: Computer-implemented methods, computer program products, and computer systems for mitigating frequency loss may include one or more processors configured for receiving first audio data corresponding to unobstructed user utterances, receiving second audio data corresponding to first obstructed user utterances, generating a frequency loss (FL) model representing frequency loss between the first audio data and the second audio data, receiving third audio data corresponding to one or more second obstructed user utterances, processing the third audio data using the FL model to generate fourth audio data corresponding to a frequency loss mitigated version of the second obstructed user utterances, and transmitting the fourth audio data to a recipient computing device. The first obstructed user utterances are obstructed by a facemask and the one or more second obstructed user utterances is obstructed by the facemask. The FL model may be executed as an audio plugin in a web conferencing program.

Abstract: A natural language processor and applicable method receives an input sentence of natural language from a natural language corpus. The input sentence comprises sentence clauses that include a conditional clause. The processor performs natural language processing (NLP), using an NLP model, on the input sentence. The processing comprises using a set of rules determining the sentence clauses and which of the sentence clause is the conditional clause, determining one or more logical connections between the sentence clauses, and determining a role of the sentence clauses based upon the one or more identified logical connections. The sentence clauses are tagged to produce a labeled sentence that is output to an entity that is one or more of a storage device, a network interface, a storage device, and an input of a further language processor or application.

Abstract: A first set of data associated with assets can be received. An ontology graph can be constructed based on the first set of data. A second set of data associated with the assets can be received, the second set of data having a first frequency of sampling. Based on the second set of data, nodes of the ontology graph representing the assets can be characterized. A third set of data associated with the assets can be received, the third set of data having a second frequency of sampling. The third set of data can include real time data associated with the assets. Based on the third set of data and information associated with the assets represented by the ontology graph, a deep learning neural network can be trained to predict a future state of at least one asset of the assets and discover dynamic mutual impact of the assets.

Abstract: Dynamically evaluating effectiveness of real-time messages is provided. A scalable signal reconstruction analysis of real-time message content in a question-and-answer format posted in a social network is performed. A message effectiveness evaluator model is generated based on the scalable signal reconstruction analysis of the real-time message content in the question-and-answer format. An effectiveness score corresponding to the real-time message content in the question-and-answer format is generated using the message effectiveness evaluator model.

Abstract: Dynamic chat discourse rephrasing is provided. An analysis is performed of a real-time chat discourse conducted between a plurality of users connected via a network. A discourse comprehension model is generated based on the analysis of the real-time chat discourse. A set of text is rephrased in the real-time chat discourse to decrease user time to comprehend the set of text using the discourse comprehension model.

Abstract: A user is detected navigating to a webpage. A plurality of web elements of the webpage is identified. One or more web elements is determined to have a risk score that surpasses a risk threshold of being a dark pattern. The risk score is identified by analyzing each of the plurality of web elements. A mitigation action configured to counteract the dark pattern is executed in response to the one or more web elements having the risk score that surpasses the risk threshold. The mitigating action is executed substantially immediately upon the user navigating to the webpage.

Abstract: A communication session between two or more computing devices is monitored. One or more share actions, that include one or more media documents, are identified in the communication session responsive to the media documents and based on the monitoring the communication session. Reaction data, that is related to the share actions, in the communication session is analyzed in response the share actions and based on the media documents. One or more relevant media documents are determined based on the analyzing the reaction data. The relevant media documents are analyzed based on the media documents. A first depiction rendered by a first of the devices of the communication session is modified based on the relevant media documents.

Abstract: Computer-implemented methods, computer program products, and computer systems for mitigating frequency loss may include one or more processors configured for receiving first audio data corresponding to unobstructed user utterances, receiving second audio data corresponding to first obstructed user utterances, generating a frequency loss (FL) model representing frequency loss between the first audio data and the second audio data, receiving third audio data corresponding to one or more second obstructed user utterances, processing the third audio data using the FL model to generate fourth audio data corresponding to a frequency loss mitigated version of the second obstructed user utterances, and transmitting the fourth audio data to a recipient computing device. The first obstructed user utterances are obstructed by a facemask and the one or more second obstructed user utterances is obstructed by the facemask. The FL model may be executed as an audio plugin in a web conferencing program.

Abstract: A computer system extracts clauses using machine translation. An input sentence in a source language is translated into a translated sentence in a target language using a trained machine translation model, wherein the trained machine translation model inserts a grammatical indicator into a position of the translated sentence that identifies a dependent clause. The input sentence and the translated sentence are aligned to determine a position in the input sentence that corresponds to the position of the grammatical indicator in the translated sentence. The dependent clause is extracted, in the source language, from the input sentence based on the determined position in the input sentence. Embodiments of the present invention further include a method and program product for clause extraction using machine translation in substantially the same manner described above.

Abstract: A computer receives reviews for an item from a plurality of sources. The computer may identify one or more key features of the item. The computer collects user preferences for the one or more key features of the item. The computer calculates odds ratio for each of the one or more key features of the item. The computer determines an affinity measure for each of the one or more key features based on the calculated odds ratio for each of the one or more key features of the item; and generates a forest plot for the item, where the forest plot comprises a summary measure determined using the affinity measure for each of the one or more key features.

Abstract: A user input is analyzed during labeling of a first portion of data to determine a behavior of a user. A labeling accuracy rate of the user is forecasted. A characteristic of a second portion of data to be presented to the user at a second time is determined. The second portion of data is presented to the user. A second label corresponding to the second portion of data is received. A second user input is analyzed to determine a second behavior of the user. The accuracy prediction model is refined. A second labeling accuracy rate of the user is forecasted. The recommendation model is refined.

Abstract: A user input is analyzed during labeling of a first portion of data to determine a behavior of a user. A labeling accuracy rate of the user is forecasted. A characteristic of a second portion of data to be presented to the user at a second time is determined. The second portion of data is presented to the user. A second label corresponding to the second portion of data is received. A second user input is analyzed to determine a second behavior of the user. The accuracy prediction model is refined. A second labeling accuracy rate of the user is forecasted. The recommendation model is refined.