38results about How to "Realize automatic mining" patented technology

An embodiment of the invention discloses a road congestion bottleneck point determination method, an apparatus, a server and a storage medium thereof. The method comprises the following steps of according to historical road condition data and road network data recorded in an electronic map service, determining a congested road section; according to the determined congested road section, and road coordinate information and historical road condition data corresponding to the congested road section, determining a road-condition space-time distribution map corresponding to the congested road section; and through a space cluster algorithm, determining a congestion cluster in the road-condition space-time distribution map and determining a space starting point of the determined congestion cluster to be a road congestion bottleneck point. In the scheme, automatic excavation of the road congestion bottleneck point is realized and accuracy of the road congestion bottleneck point is increased.

The invention provides a method and a device for excavating a badcase (badcase) of a search engine, wherein the method comprises the following steps of a preprocessing procedure: extracting a certain number of sessions as samples from a session (session) log, and extracting a feature vector describing the search quality from each session of the samples; clustering the samples by utilizing the feature vector of each session; determining confidence coefficient of each category obtained by clustering the samples, wherein the confidence coefficient represents the low degree of the search quality; an excavating procedure: determining an action sequence in the same query in a session log to be excavated, and extracting a feature vector describing the search quality from the action sequence; determining the category of the query by computing the distance between the feature vector of the query and the feature vector of each category; if the confidence coefficient of the category of the query is beyond a preset high threshold, determining that the search engine has the badcase to the query. According to the method and device for excavating the badcase of the search engine, which are disclosed by the invention, the automatic excavation of the badcase of the search engine can be realized, so that the badcase of the search engine can be timely and exactly found out.

The present application discloses a negative sample mining method, device, electronic device and storage medium, and relates to the fields of artificial intelligence such as deep learning and natural language processing. The method may include: when using at least two graphics processors to perform multi-machine model training, Any graphics processor performs the following processing: in any training batch, each query in the training batch is regarded as a pending query; for each pending query, the other queries in the training batch except the pending query are respectively Each query and each query in the training batch corresponding to each graphics processor except itself are used as an associated query, and the sample corresponding to each associated query is used as a weak negative sample corresponding to the query to be processed, and the corresponding samples include positive samples and / or or strong negative samples. By applying the solution described in this application, the model training effect and data utilization rate can be improved.

The invention provides a rail electric locomotive automatic transportation system based on 5G technology. The system comprises an automatic transportation electric locomotive, an electric locomotive control system, 5G signal receiving and converting equipment CPE, a pantograph, a mine car, a 5G antenna, a 5G transmission feeder line, a 5G base station RRU, a computer terminal, a turnout control PLC substation and a synchronous electromagnet. The electric locomotive is connected to the mine car, and is mainly used for traction of the mine car. An electric locomotive control system is fixedly mounted in an electric locomotive cab, a PLC is arranged in the electric locomotive control system and used for processing the electric locomotive position, electric locomotive running speed, electric locomotive running mode, electric locomotive running current, electric locomotive frame line voltage and other data, so that intelligent running of the underground electric locomotive is achieved, andthe running efficiency of the electric locomotive is improved.

The invention provides an automatic mining method of a requirement identification template, a requirement identification method and a corresponding device. The automatic mining method comprises the following steps of: determining a corresponding query set when a webpage of a preset type is clicked in a search log; selecting a query that the total clicking number of the webpage of the corresponding preset type exceeds a preset frequency threshold and / or the clicking rate of the webpage of the corresponding preset type exceeds the preset clicking rate threshold from the query set; taking the selected query as a seed query of the preset type; matching each seed query with a dictionary of the preset type respectively; substituting words in the seed query matched with the dictionary into attribute marks of the corresponding words in the dictionary to obtain a template set of the preset type; and determining the requirement identification template of the preset type by using the template set of the preset type. According to the automatic mining method, the manpower resource can be saved, the query range capable of being covered by search identification is expanded, and the recall rate is improved.

The present invention relates to a method, system, device and medium for generating addresses of buildings in a target area based on spatial analysis. Name feature words and sequential feature words are extracted from address names of known buildings in the target area, wherein The building bodies of the feature words are the same redundant data set; in the same redundant data set, according to the preset spatial relationship between the building face vector data of the sequential feature words meeting the preset requirements and the building face vector data to be identified The address name that identifies the building. According to the preset spatial relationship between the buildings in the target area, the name feature words and sequence feature words in the known building address names are used to determine the name feature words and sequence feature words of the buildings to be identified, which reduces the work of on-site collection The cost is reduced, the operation efficiency is improved, the accuracy of navigation electronic map positioning and navigation is greatly improved, and the user experience satisfaction is ensured.