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 discloses a freight trajectory data-based vehicle spontaneous formation mode mining method, and the method comprises the steps: completing the freight spontaneous formation mode search based on a data mining method according to the daily massive freight trajectory data; providing a full-process processing framework innovatively; setting seven databases, using an improved running distance estimation method, a map matching method, a P-OPTICS algorithm and the like to complete tasks of estimating the freight travel of each truck on that day, truck travel positioning point map matching and coordinate repairing, hotspot line and excavation time period planning, truck formation set excavation at each moment, truck spontaneous formation mode excavation and evaluation under a specified line and a specified time period and the like. Therefore, key information such as the running distance is extracted from massive freight trajectory data, and meanwhile, a freight car spontaneous formation mode mining task in a specified road section or time period is completed, so that business reference is provided for freight car formation driving work under an automatic driving background in the future while formation mode characteristics are analyzed.

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 invention provides a public opinion analysis method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a plurality of pieces of public opinion information corresponding to each target business in at least one target business; for each target business, determining public opinion graph data of the target business based on a pre-constructed knowledge graph structure and multiple pieces of public opinion information corresponding to the target business; and constructing a knowledge graph database corresponding to each target service according tothe public opinion graph data so as to perform association analysis on at least one target service by utilizing each knowledge graph database. By adopting the scheme, the public opinion graph data inthe public opinion information is extracted in combination with the knowledge graph structure, so that association analysis among the target services is realized through the constructed knowledge graph database, automatic mining of the public opinion information is realized, and the public opinion analysis efficiency is improved.

Embodiments of the present invention provide a data processing method, device and server. The method comprises: acquiring user features of users to be excavated from at least one data source; acquiring a pre-trained target asset status predicting model which is obtained by training according to the user features of positive sample users and negative sample users acquired from the at least one datasource, wherein the possibility that the positive sample users have target assets is larger than the possibility that the negative sample users have target assets and the user features at least comprise user behavior features; predicting the probability that the users to be excavated have the target assents according to the user features of the users to be excavated and the target asset status predicting model; and determining that the users to be excavated have the target assents if the probability that the users to be excavated have the target assents is larger than a probability threshold.The embodiments of the present invention can increase processing efficiency of excavation of a user asset status.

The invention discloses a vocabulary mining method and apparatus. The method comprises the steps of determining an entity word set and a candidate hypernym set comprised in a corpus sentence; combining words in the two sets in pairs to obtain candidate word pairs; further determining word vectors of the entity words and the candidate hypernyms in the candidate word pairs; and according to the wordvectors, determining whether the candidate word pairs are vocabulary mining results or not. An example is as shown in that whether the candidate word pairs are hypernym pairs or not is determined. Corpora do not need to be manually sorted; and automatic mining of the hypernym pairs is realized by machine learning, so that the hypernym pair mining efficiency is greatly improved and the mining costis reduced.

The invention provides an ultrasonic examination follow-up patient screening method based on machine learning. Due to the rapid development of the deep learning technology, the utilization of the natural language processing technology and the deep learning technology becomes an important means for analyzing medical texts, and the means is an effective way for replacing manual text screening. According to the method, word segmentation is carried out on text contents through a JIEBA word segmentation tool, a TF-IDF method and a Word2Vec algorithm are adopted to construct word vectors respectively, and a chi-square test method is further utilized to select feature vectors; and a classification model selects XGBoost, Lightgbm and CNN to perform training modeling on feature data, so automatic screening of an ultrasonic examination follow-up list is realized.

The invention provides a data processing method and device. The method includes: determining feature vector words of words to be processed; using internal chain words set and occurring in a results page which the words to be processed specially belong to, as candidate internal self-chain words of the words to be processed; performing calculation according to a set recommended score calculating method, and calculating a recommendation score of each candidate internal self-chain word through a feature vector word of each candidate internal self-chain word and the feature vector words of the words to be processed; selecting a set number of candidate internal self-chain words having high recommendation scores, as internal self-chain words related to the words to be processed. The data processing method has the advantage that an internal self-chain word of a word can be automatically mined during processing of the word.

The invention provides a search result processing method and device and a server, and belongs to the technical field of computers. The search result processing method comprises the following steps: after acquiring search terms, search frequency of the search terms and feature information of a plurality of search results corresponding to the search terms, acquiring the effective value of each search result according to the search frequency and the feature information; constructing a first search result sequence which corresponds to the search terms according to the effective value of every search result; constructing a second search result sequence corresponding to the search terms according to the showing position of every search result in search pages corresponding to the search terms; and comparing the first search result sequence with the second search result sequence to obtain bad examples in a plurality of search results. By the search result processing method, the bad examples inthe search results corresponding to the search terms can be automatically mined, a large amount of labor is saved, and the bad example mining efficiency is improved.

The invention discloses an event mining method and device. According to the method, a plurality of reference texts can be extracted from a text data set; event classification is carried out on the reference texts to obtain an event text set, wherein the event text set comprises reference texts associated with a target event; clustering processing is carried out on reference texts in the event textset to obtain a clustered text set corresponding to sub-events, wherein the sub-events belong to the target event; a mapping relationship between the text quantity of a target text in the text data set and the search time of the target text is analyzed to obtain an analysis result, wherein the target text is a text belonging to the sub-event in the text data set; and periodic discrimination is performed on the sub-events based on the analysis result, and a periodic discrimination result is output. According to the method, automatic mining of periodic events can be realized, the accuracy of event mining results is improved, and the mining efficiency can also be improved.

The invention discloses an automatic insight discovery method for multi-dimensional time series data. The method comprises the steps of S1, specifying a data set and hyper-parameters by a user; S2, searching the insight with the maximum insight score; and S3, visualizing the searched insights and displaying the insights to the user. According to the method and the device, the technical problems that manpower and time costs are consumed and manual operation and judgment are excessively depended when the multi-dimensional time sequence data is manually analyzed are solved, automatic mining of insight of the multi-dimensional time sequence is realized, and the searching efficiency is improved.

The invention relates to the related field of mining and processing of mineral resources, in particular to automatic coal collecting and classifying equipment which comprises a box body and a first cavity fixedly formed in the box body, wherein wheel shafts are rotationally mounted on the end wall of the lower side of the first cavity in a bilateral symmetrical mode, wheel shafts are fixedly mounted on the outer surfaces of the tail ends of the front side and the rear side of each wheel shaft, the left wheel shaft is in power connection with a spindle of the motor, and the motor is fixedly installed on the side wall of the lower side of the first cavity. According to the automatic coal collecting and classifying equipment, automatic mining of coal resources can be achieved, workers do notneed to enter a mine for mining, once an accident occurs, life health of the workers is not threatened, the mined coal resources can be placed in a classified mode, no special equipment is needed forclassifying the mined coal resources, the sizes of coal briquettes can be distinguished through a rotating baffle, and then different sorting devices are driven to be placed in a classified mode.

The invention discloses a map interest point mining method, belonging to the field of big data and the field of natural language processing. The method comprises the steps of collecting at least one piece of Wi-Fi scanning information and positioning information of multiple pieces of user equipment; according to the at least one piece of Wi-Fi scanning information and the positioning information of the plurality of pieces of user equipment, acquiring associated information of at least one Wi-Fi, wherein the associated information comprises a name and a geographic position; if the name of the Wi-Fi is the name of the interest point, adding a corresponding map interest point to an electronic map according to the name and the geographic position of the Wi-Fi, thereby effectively improving themining efficiency of the map interest point. The invention further discloses a map interest point mining device, equipment and a storage medium.

The invention relates to the field of engineering machines, and discloses a control method and device for an engineering machine, a processor, and the engineering machine. The control method comprises the steps: receiving a first working parameter of a first working task; determining whether a first operation instruction set corresponding to the first working parameter is present in a pre-stored database according to the first working parameter; when it is determined that the first operation instruction set is present, obtaining the first operation instruction set, and controlling the engineering machine to execute the first working task according to the first operation instruction set; and under the condition that it is determined that no first operation instruction set is present, obtaining a second operation instruction set corresponding to a second working parameter closest to the first working parameter from the database, and controlling the engineering machine to execute the first working task according to the second operation instruction set. According to the technical scheme, the operation instruction sets in the expert database can be applied, automatic work of the engineering machine is achieved, labor cost in work is reduced, and working precision and working efficiency are improved.

The invention relates to a method, a system, a device and a medium for generating a building address in a target area based on spatial analysis, wherein, a name feature word and a sequential feature word are extracted from address names of known buildings in the target area, wherein, the buildings with the same name feature word are the same redundant data set; In the same redundant data set, theaddress names of the buildings to be identified are generated according to the preset spatial relationship between the building surface vector data and the building surface vector data to be identified which satisfy the preset requirements by the sequential feature words. According to the preset spatial relationship among the buildings in the target area, The name feature words and the sequence feature words of the known building address names are used to determine the name feature words and the sequence feature words of the building to be recognized, which reduces the operation cost of the on-site collection, improves the operation efficiency, greatly improves the positioning and navigation accuracy of the navigation electronic map, and ensures the satisfaction of the user experience.