Abstract: Provided is a method for embedding a knowledge graph. The method includes: obtaining a basic knowledge instance and a qualifier instance associated with the basic knowledge instance, wherein at least one of the basic knowledge instance or the qualifier instance comprises a mask element; generating, through an embedder, a first instance embedding corresponding to the basic knowledge instance and a second instance embedding corresponding to the qualifier instance; and performing a prediction of the mask element by inputting, to a predictor, an instance embedding associated with the mask element, wherein the instance embedding corresponds to the first instance embedding or the second instance embedding.

Abstract: A method and device for providing a closed messenger service to a second authority-level user includes receiving, from a first terminal, a first identity of a first authority-level user who invited the second authority-level user, a second identity of the second authority-level user, and information identifying a workspace to which the second authority-level user is invited, transmitting an authorization request to an approver when it is determined that authorization for the second authority-level user is required on the basis of the first identity, and performing a workspace member registration procedure based on the second identity and the information identifying the workspace when receiving an authorization confirmation to the authorization request from a second terminal.

Abstract: An announcement display method according to an embodiment of the present disclosure is performed by a computing device, and includes displaying an intro screen including a friend account list of a user account, and fixedly displaying an announcement list at the top of the intro screen, wherein the announcement list comprises an announcement of at least one chat room to which the user account belongs.

Abstract: A device for retrieving previous chat history from a chat room includes a control module configured to generate a chat history retrieval request, a transmitting/receiving module configured to transmit the generated chat history retrieval request to a server, and to receive, from the server, a previous chat history, and an input/output module configured to display the received previous chat history in a chat window. The invitation message at least includes an ID of the first user, an ID of the second user, an ID of the chat room, a transmission time of the invitation message, and a previous message code. The previous message code is unit time for limiting the retrieval start point.

Abstract: Provided is a method performed in a terminal for processing an interest message. The method includes receiving a message from a message server after a login event for a user of the terminal occurs, wherein the login event is related to a message service provided by the message server; storing, in response to determining that the received message corresponds to an interest message of the user, the received message in a local storage of the terminal; receiving a search request for the interest message of the user; and searching and providing the interest message of the user through the local storage of the terminal in response to the search request.

Abstract: A receiving device for processing an attention message, the receiving device includes a transmitting/receiving module configured to receive a message, a control module configured to determine whether the received message is an attention message, and to set an attention display flag, and an input/output module configured to display the attention indicator in the chat room in response to the set attention display flag. The message includes an ID of the chat room, an ID of the transmitting device, an ID of the first user, a type of the message, a content of the message, and a transmission time. The control module is configured to determine, based on at least one of the type of the message, the content of the message, and the transmission time, whether the message is the attention message.

Abstract: Provided is a method performed by a least one computing device for a message service. The method comprises receiving a transmission request for a compound message which is a bundled message comprising a parent message and a plurality of child messages, obtaining permission information for a sender of the compound message, determining whether to allow each of the child messages to be sent based on the permission information and controlling transmission of the compound message based on the determination result.

Abstract: Provided is a method for a token verification performed by a service server. The method according to some embodiments of the present disclosure comprises receiving a service request message comprising a token from a client, performing a first verification process for the token through local storage of the service server in response to the service request message and performing a second verification process for the token through another service server related to a used service list in response to failure of the first verification process, wherein the service request message further comprises the used service list of the client.

Abstract: An apparatus for embedding a sentence feature vector according to an embodiment includes a sentence acquisitor configured to acquire a first sentence and a second sentence, each including one or more words; a vector extractor configured to extract a first feature vector corresponding to the first sentence and a second feature vector corresponding to the second sentence by independently inputting each of the first sentence and the second sentence into a feature extraction network; and a vector compressor configured to compress the first feature vector and the second feature vector into a first compressed vector and a second compressed vector, respectively, by independently inputting each of the first feature vector and the second feature vector into a convolutional neural network (CNN)-based vector compression network.

Abstract: Provided are a method for evaluating performance and a system thereof. The method according to some embodiments may include obtaining a first model trained using a labeled dataset, obtaining a second model built by performing unsupervised domain adaptation on the first model, generating pseudo labels for an evaluation dataset using the second model, wherein the evaluation dataset is an unlabeled dataset, and evaluating performance of the first model using the pseudo labels.

Abstract: Provided are a method for domain adaptation and a system thereof. The method according to some embodiments of the present disclosure may include obtaining a model trained using a labeled dataset of a source domain, wherein the model comprises a feature extractor configured to extract a feature from an input data sample and a predictor configured to predict a label based on the extracted feature, selecting a first data sample from an unlabeled dataset of a target domain, generating a second data sample through data augmentation on the first data sample, calculating a consistency loss between the first data sample and the second data sample by using at least one of the feature extractor or the predictor, and updating the feature extractor based on the consistency loss.

Abstract: In a method of generating a secret key according to an embodiment, a share of each of a user and a plurality of other users for a secret key of the user are generated, the share of each of the plurality of other users is provided to a user terminal of each of the plurality of other users, a share of the user for a secret key of each of the plurality of other users is received from the user terminal of each of the plurality of other users, and a new secret key of the user is generated using the share of the user for the secret key of the user and the shares of the user for the secret key of each of the plurality of other users.

Abstract: A device for retrieving previous chat history from a chat room includes a control module configured to generate a chat history retrieval request, a transmitting/receiving module configured to transmit the generated chat history retrieval request to a server, and to receive, from the server, a previous chat history, and an input/output module configured to display the received previous chat history in a chat window. The invitation message at least includes an ID of the first user, an ID of the second user, an ID of the chat room, a transmission time of the invitation message, and a previous message code. The previous message code is unit time for limiting the retrieval start point.

Abstract: A method for extracting an area of interest in a document is provided. The method may comprise extracting one or more target pages from a document composed of a plurality of pages and extracting an area of interest including a plurality of sentences from a target page based on a first part-of-speech characteristic and a sentence characteristic of the target page.

Abstract: A method for protecting data based on a private set union (PSU) protocol includes: generating a Boolean vector for a first information set, based on whether or not each of a plurality of first groups that are groups of elements of the first information set stored in a first computing device is the same as each of a plurality of second groups that are groups of elements of a second information set stored in a second computing device, obtaining a random vector for the first information set as a result of shuffling performed by using the Boolean vector generated by the first computing device as an input and using order information and encryption information selected by the second computing device as inputs, and generating information on a union of the first information set and the second information set based on the random vector.

Abstract: A device for storing cylindrical material according to an embodiment of the present disclosure includes a rack shelf including one or more racks in which a circular axis is stored, a rack master including a first rack master fork configured to collect a first circular axis in which an article stored in a target rack is not mounted, and a second rack master fork configured to transport a second circular axis in which the article is mounted in the target rack from which the first circular axis is collected, a turn skid configured to collect the first circular axis collected from the first rack master fork; and an input/output port configure to mount the article in the first circular axis collected from the turn skid.

Abstract: A semantic segmentation method performed by at least one computing device is provided. The method may include extracting a plurality of support features, which include a first-level feature and a second-level feature, from a support image; generating a plurality of prototypes for a class associated with the support image, based on the plurality of support features; and predicting a segmentation mask for a query image using the plurality of prototypes, wherein the plurality of prototypes comprise a first prototype generated based on the first-level feature and a second prototype generated based on the second-level feature.

Abstract: Provided is a embedding transformation method performed by at least one computing device. The method comprises obtaining a source-side embedding model, transforming source-side data into a first embedding vector through the source-side embedding model and transforming the first embedding vector into a second embedding vector located in a target-side embedding space through a transformation model.

Abstract: An apparatus for analyzing a network according to an embodiment includes a receiving module configured to receive, from a cloud service, information on one or more resources included in the cloud service and information on a network environment of the one or more resources, a topology generation module configured to derive a network topology of the cloud service by using the information on the resources and the information on the network environment, and an analysis module configured to derive, from the network topology, a communication possible path of each of the one or more resources and one or more communication allow policies in the communication possible path.

Abstract: A method and an apparatus for creating a workflow, the method being performed by a computing device, and including obtaining a log file generated while a user performs a task, parsing each of a plurality of events recorded in the log file, and creating an activity sequence to correspond to an order of the events, grouping a plurality of partial sequences extracted from the activity sequence, and creating a plurality of activity groups to correspond to each of a plurality of different repetitive tasks included in the task and creating the workflow of at least one activity group among the plurality of activity groups.