-
公开(公告)号:US20190087724A1
公开(公告)日:2019-03-21
申请号:US15856463
申请日:2017-12-28
Inventor: Young Tack PARK , Wan Gon LEE , Batselem JAGVARAL , Hyun Young CHOI , Ji Houn HONG
Abstract: Disclosed are a method of operating a knowledgebase and a server using the same. The method of operating a knowledgebase may include: (a) receiving triple data sets as input; (b) forming at least one data cluster set by classifying the triple data sets according to relation based on semantic information of the knowledgebase; and (c) learning each relation model by inputting the data cluster set into a neural tensor network.
-
公开(公告)号:US20210064935A1
公开(公告)日:2021-03-04
申请号:US16697238
申请日:2019-11-27
Inventor: Young Tack PARK , Jagvaral BATSELEM , Wan Gon LEE
Abstract: A triple verification method is provided. The triple verification method includes setting a triple having a source entity, a target entity, and a relation value between the source entity and the target entity by a setting unit, extracting a plurality of intermediate entities associated with the source entity and the target entity by the setting unit, defining a connection relation between the intermediate entity, the source entity, and the target entity and generating a plurality of connection paths connecting the source entity, the intermediate entity, and the target entity by a path generation unit, generating a matrix by embedding the plurality of connection paths into vector values by a first processing unit, calculating a feature map by performing a convolution operation on the matrix by a second processing unit, generating an encoding vector for each connection path by encoding the feature map by applying a bidirectional long short-term memory neural network (BiLSTM) technique by a third processing unit, and generating a state vector by summing the encoding vectors for each connection path by applying an attention mechanism and verifying the triple based on a similarity value between the relation value of the triple and the state vector by a determination unit.
-
3.
公开(公告)号:US20190138925A1
公开(公告)日:2019-05-09
申请号:US16170460
申请日:2018-10-25
Inventor: Young Tack PARK , Myung Joong JEON , Seok Hyun BAE , Je Min KIM , Hyun Kyu PARK , Sung Hyuk BANG
Abstract: A terminal device for generating user behavior data, a method for generating user behavior data, and a recording medium are provided. The disclosed terminal device may include a memory unit storing instructions readable by a computer; and a processor unit implemented to execute the instructions, where the processor unit may compute a probability distribution model for achieving the intentions of a user by using raw data related to time-dependent actions of the user and may generate user behavior data by using the probability distribution model, with the user behavior data comprising time series data in which multiple actions composing the intentions of the user are aligned in order.
-
公开(公告)号:US20200175226A1
公开(公告)日:2020-06-04
申请号:US16696476
申请日:2019-11-26
Inventor: Young Tack PARK , Wan Gon LEE , Jagvaral BATSELEM , Hyun Young CHOI , Ji Houn HONG
IPC: G06F40/20 , G06F40/166 , G06K9/62 , G06N20/10 , G06N20/20
Abstract: Provided is an incorrect triple detection system including a triple selector configured to select a target triple (subject, type, object) in a knowledge base, a sampler configured to create a sentence model by connecting object triples sharing entities included in the target triple, a model builder configured to embed the sentence model into a vector space to create a training entity vector and build an embedding model, and an incorrect triple detector configured to detect an incorrect triple by inputting a test triple into the embedding model.
-
公开(公告)号:US20190087732A1
公开(公告)日:2019-03-21
申请号:US15852576
申请日:2017-12-22
Inventor: Young Tack PARK , Wan Gon LEE , Sung Hyuk BANG
Abstract: Disclosed are ontology-based reasoning apparatus, system, and method using user-defined rules. The ontology-based reasoning apparatus includes a memory storing two or more triple data; and a processor connected to the memory, in which the processor parses user-defined rules input from a user in ontology-based rules to generate a plurality of condition meta information corresponding to a plurality of conditions included in a condition section of the user-defined rules and one result meta information corresponding to one result included in a conclusion section of the user-defined rules, retrieves a plurality of triple data in two or more triple data prestored using the plurality of condition meta information, and reasons new triple data from the plurality of triple data retrieved using the result meta information to store the reasoned new triple data in the memory.
-
公开(公告)号:US20170316327A1
公开(公告)日:2017-11-02
申请号:US15281959
申请日:2016-09-30
Inventor: Young Tack PARK , Hyun Kyu PARK , Wan Gon LEE , Batselem Jagvaral
CPC classification number: G06N5/022
Abstract: Ontology reasoning method and apparatus are disclosed. The ontology reasoning method comprises (a) broadcasting a schema triple to nodes; (b) partitioning triples other than the schema triple and distributing the partitioned triples to the nodes; and (c) reasoning at least one of the schema triple or the triples according to a reasoning rule and then renewing a confidence value of the reasoned triple.
-
公开(公告)号:US20170316315A1
公开(公告)日:2017-11-02
申请号:US15470886
申请日:2017-03-27
Inventor: Young Tack PARK , Hyun Kyu PARK
IPC: G06N5/02 , G06F17/30 , G06N5/04 , G06F3/0481
CPC classification number: G06F16/90335 , G06N5/045
Abstract: Ontology-based reasoning apparatus and method using knowledge of an expert are disclosed. The ontology-based reasoning apparatus includes an input unit through which a plurality of the knowledge of the expert including a condition and a result are inputted, a conversion unit configured to convert the inputted knowledge to ontology-based rules, and a reasoning performing unit configured to reason a conclusion by using the converted rules and generate an explanation of a rule corresponding to cause of the conclusion. Here, the rule includes a condition node corresponding to the condition, a result node corresponding to the result and an edge for connecting the condition node to the result node, the edge includes an edge direction and an edge value, and the edge value corresponds to an explanation about relation between the condition node and the result node.
-
-
-
-
-
-