47 results about "Semantic grid" patented technology

A computer implemented method and computer usable program product for version control of source code. In one embodiment, a source code file is scanned for relationships between source code sections. A semantic graph with dependencies and metadata describing the source code sections is constructed. The dependencies indicate the relationships between the source code sections. If changes are received for the semantic graph, the semantic graph is modified with the changes to form a modified semantic graph.

The invention relates to a method for generating a semantic grid map and a device thereof, wherein the method comprises the following steps of: acquiring image data of an environment through a visualsensor, acquiring 2D scanning data of the environment through a laser radar sensor; merging the acquired image data and the 2D scanning data to generate semantic scanning data; merging the generated semantic scanning data with the grid map of the environment to generate a semantic grid map. According to the invention, the visual sensor is introduced, the image data is acquired through the visual sensor, and the semantic information is endowed to the 2D scanning data acquired through the laser radar sensor; during the merging process of the map, the dynamic object is removed to reduce the noisepoint, meanwhile, the semantic information is added to the determined grid, the semantic grid map is generated, and reliable road sign information is provided for the composition, positioning and navigation, thereby improving the robustness of the composition and positioning of a mobile robot in the dynamic environment.

Non-hierarchical relationship UIs (“topology UIs”) are difficult to scale to hundreds or more items. An embodiment of this invention provides indicators and quick access controls for the nodes in navigation tree that appear more than once in a hierarchy—where a pure hierarchical navigational tree falls short. The navigational tree is structured by grouping all matching nodes together. The display area displays the most frequent and the concatenated labels of the semantically grouped redundant nodes. A set of contextual proximity and similarity information and an exception list are stored in a semantic database which is configurable by system administrators. These navigation area controls and indicators will preemptively handle non-hierarchies directly within the navigation tree.

The invention relates to a semantic index method based on domain ontology, which comprises steps of (1) using singular value decomposition to decompose term-document matrices to obtain concept term-document matrices, (2) firstly establishing a basic ontology repository of the domain and using the basic ontology repository to construct a dynamic semantic template library based on the domain ontology, (3) establishing an inverted index for the document by using the concept terms which are added to the dynamic semantic template library in step (2) by a semantic index which is a secondary index undertaken on the base of the ontology and relevant statement thereof, and simultaneously establishing secondary indexes for the relevant statements appeared under each ontology. The invention has the advantages that the shortage that the keyword index is difficult to overcome and identify synonyms, semantically same or associated information can be obtained, and the secondary index of the relevant statement of the ontology can refine search granularity, can provide higher accuracy and efficiency for search results, and can provide advanced search functions for users.

The invention discloses a short text classification method based on semantic graphs, which is characterized by including the steps: A, constructing a semantic graph model for each piece of text information, and combining all semantic graph models; B, comparing similarity level among different texts according to the semantic graph models and by means of a similarity computing method; and C, according to the text similarity level, classifying the texts by the aid of a text semantic graph classifier. The short text classification method based on the semantic graphs has the advantages that semantic connotations of documents can be highlighted to a maximum degree by using the graph models to represent the texts, latent semantic information and theme features in the texts can be accurately described to a great extent by the aid of the TSG (text semantic graph) models constructed by the method, and the TSG classification method can be more reliable and efficient in use as compared with other classification methods by means of the feature, so that human cost is greatly reduced, artificial arrangement of the text information is avoided to a great extent, and the text information is automatically organized by a computer.

The invention discloses a semantic risk calculating method based on a text logical characteristic. The semantic risk calculating method comprises the steps of predefining a structure type, and forming a preset-type semantic logic combination; dividing a to-be-processed text into a plurality of texts, and performing word cutting; for the word combination of the to-be-processed paragraph, performing matching according to a predefined preset-type semantic logical combination, performing traversal of the preset-type semantic logical combination, performing paragraph risk value calculation on all word sets with defined semantic structures; performing matching calculation processing on all paragraphs; classifying matching risk sets of all paragraphs of the to-be-processed text, performing accumulation according to the paragraph weight and the paragraph risk value, and obtaining sequential risks and scores of the risks after protocol calculation. The semantic risk calculating method can realize high-efficiency and accurate risk mining.

The invention discloses a method and a system for acquiring answers of tax business questions based on semantic graphs, and belongs to the technical field of intelligent questions and answers. The method comprises the steps of defining a question Q and an answer of the question Q according to a state transition graph G, and mapping a semantic interpretation process of the question Q into a sub-graph sub-G query process of the state transition graph G; generating a logic expression; action transfer and conditions in the logic expression are repeated; and obtaining a tax business problem, inputting the tax business problem into the model, checking whether the tax business problem completes the logic expression X, and outputting a tax business problem answer when the tax business problem completes the logic expression X. The semantic parsing process is mapped into the retrieval process of the query graph, so that the retrieval space of problem parsing is reduced, the operation between entities is judged by using the deep convolutional neural network, and the predicate synonym ambiguity problem mentioned in the preamble is solved.

The invention discloses a text recommendation method based on depth semantic discrimination. A text topic is automatically extracted according to a depth semantic grid model, scene semantics under different text backgrounds are inferred according to a topic scene semantics analysis method, a text topic tree which incorporates situational states is achieved, a user text interest portrait is constructed for each document based on a user's real-time situational state. At the query end, for the real-time fluctuations of the user's situational state, the text topic tree is screened for contextual semantics, the query content is modeled on the query interest topic, according to an activation diffusion method, second latent semantic reasoning is performed on the user's direct interest topic, theglobal activation value of the topic is calculated, and a user query interest portrait which incorporates the current context semantics is constructed. The documents are scored by a similarity calculation method, and a text recommendation list is generated according to the score.

An electronic device for determining a semantic grid of an environment of a vehicle is provided. The electronic device is configured to: receive first image data of an optical sensor, the first image data comprising a 2D image of the environment, perform a semantic segmentation of the 2D image and project the resulting semantic image into at least one predetermined semantic plane, receive an occupancy grid representing an allocentric bird eye's view of the environment. The control device further includes a neural network configured to determine a semantic grid by fusing the occupancy grid with the at least one predetermined semantic plane.

The invention discloses an improved AMCL positioning method based on a semantic map with wall corner information and a robot. According to the method, an established two-dimensional semantic grid map is utilized, and the semantic map is fused with a target detection method based on deep learning to extract central point positions of wall corners and other various objects in the environment in the grid map, so that a semantic information graph around the corner points is formed. And near obstacle distance value calculation is performed on the loaded grid map, and a semantic lookup table around the angular points is generated. And a proposed visual pre-positioning method is combined, so that the mobile robot can realize initial positioning more quickly under the condition of a small amount of prior information and movement. And a particle weight updating mode is improved, and an AMCL algorithm and environment map matching are synchronously combined to carry out fine positioning. Contrast experiments are carried out under different environment conditions, and it is verified that the method improves the particle convergence rate of the robot and particularly improves the positioning accuracy and robustness when the environment has certain similarity or changes to a certain extent.

A dataset including boundary representations of shapes associated with an item being designed for manufacture is accessed by a semantic processing module. A semantic graph of each of the boundary representations of shapes is generated and a numerical processing module computes geometric attributes of each of the shapes. The semantic graph of a shape is updated based on any geometric attributes computed for the shape. The semantic graphs are then compared to a repository of semantic graphs of manufacturing features to identify instances of manufacturing features. Geometric attributes associated with each instance of a manufacturing feature are then computed. For each instance, the associated geometric attributes are compared against a repository of semantic manufacturing rules to determine whether the instance is in compliance with the rules. A user designing the item is alerted to the presence of any instances of manufacturing features that are not in compliance with the rules.

The invention relates to a Chinese medical archeography database. It stores special topic database, original domicile document database, index database and original domicile image database into the computer; the original domicile document database and the index database are graded by the bibliography principle, and the bibliography contents with different grades are stored into the corresponding grade fields, the content is stored on the last grade by section type. The established Chinese medical archeography database possesses the function of editing, grading and storing, the entire index, rear controlling and semantics index.

The embodiment of the invention provides a positioning method and device, electronic equipment and a storage medium. The method comprises the steps that a target local semantic point cloud map, a global semantic grid map and a laser point cloud map of a current position are acquired, and each grid of the global semantic grid map records a probability value that a target semantic object exists in the grid; pose recognition is performed based on the target local semantic point cloud map and the global semantic grid map to obtain a candidate pose set; and a target pose is determined according to the laser point cloud map and the candidate pose set, wherein the target pose comprises a loopback pose and/or a relocation pose. According to the embodiment of the invention, by adopting the technical scheme, the calculation amount required in the positioning process can be reduced, and the loopback detection/repositioning speed can be improved.

The invention provides a visualization concept detector which is used for a retrieval process of a three-dimensional model. The visualization concept detector comprises a text description module, a semantic description module, a feature description module and a semantic concept computation module, wherein the text description module, the semantic description module and the feature description module are respectively used for constructing text description, network annotation and semantic concepts of synonym sets and shape and content characteristics in order to obtain a semantic concept vocabulary, and the text description, the network annotation and the semantic concepts are originated from a to-be-retrieved three-dimensional model. The semantic concept computation module is used for calculating through a probability model to obtain a core semantic concept vocabulary, namely a semantic field, and the visualization concept detector is combined to retrieve the three-dimensional model. Therefore, the harmful effect of 'semantic gap' is weakened, the calculated amount of three-dimensional model retrieval is reduced, the retrieval speed is accelerated, and the precision and the stability of retrieved results are enhanced.

Building a semantics graph for an enterprise communication network can include calculating a distance metric between a first signifier and a second signifier associated with an enterprise communication network, wherein the distance metric includes a plurality of relationships defined based on a frequency of co-occurrences of the first signifier and the second signifier, and building a semantics graph for the enterprise communication network using the calculated distance metric.

The invention provides a method and system for adjusting a corpus disordered in semantic logic. The method includes the steps that a corpus sample clear in logic and complete in semantics is obtained,and according to the corpus sample, a voice library, a semantic slot and a regular expression library are built; user voice is acquired; the user voice is matched with the voice library, and a matched word segment is obtained, wherein the matched word segment is a word segment conforming to a matching result in the user voice; matched word segment part-of-speech corresponding to the matched wordsegment is determined according to the semantic slot; the position of the word segment in the user voice is adjusted according to a regular expression in the regular expression library and the matchedword segment part-of-speech, so that text data with correct logic is obtained; semantic analysis is carried out according to the text data. According to the method, the relative position between theword segments in the corpus disordered in semantic logic is adjusted, so that the real user intention is intelligently recognized.