171results about How to "Fix technical issues with lower accuracy" patented technology

The invention provides an advertisement ordering method comprising the steps that advertisement image characteristic parameters and advertisement text characteristic parameters are acquired; the advertisement image characteristic parameters and the advertisement text characteristic parameters are set as input vectors of a first depth neural network; first model training is performed on the first depth neural network so that pre-estimated advertisement click-through rate characteristics are outputted; and advertisement ordering is performed according to the pre-estimated advertisement click-through rate characteristics. The invention also provides an advertisement ordering device. According to the advertisement ordering method and device, statistics is performed on the characteristics of advertisement images via the convolutional neural network so that accuracy of advertisement click-through rate pre-estimation is enhanced.

The invention discloses a method and device for detecting network attacks. The method comprises the following steps: receiving a network request from a sender; when determining that the type of the network request is an attack request by adopting an attack detection rule set, forwarding the network request to a receiver, and acquiring a network response corresponding to the network request; and detecting the network response by adopting the attack detection rule set, and selecting a processing mode of the network response according to a detection result. By adopting the method and device disclosed by the invention, the technical problem that a one-way network attack detection method adopted in related technologies is relatively low in accuracy can be solved.

The invention discloses a method, device and terminal for detecting game plug-in. The method comprises the following steps of: acquiring data to be detected of an object to be detected; detecting datato be detected based on a plurality of unsupervised learning models to obtain a plurality of characterization vectors corresponding to the plurality of unsupervised learning models, wherein the plurality of unsupervised learning models are used for training data of a plurality of objects to obtain a plurality of characterization vectors; carrying out clustering processing on a plurality of characterization vectors from a preset dimension to obtain a clustering result; obtaining the doubt degree corresponding to the preset dimension according to the clustering result; and judging whether the object to be detected is an abnormal object or not according to the doubt degree corresponding to the preset dimension, wherein the abnormal object has a plug-in behavior. The method, device and terminal for detecting game plug-in solves the technical problem of low accuracy of detecting the game plug-in in the prior art.

The invention provides an accompanying object determining method and device and an electronic device, which relate to the technical field of data mining. The accompanying object determining method comprises the following steps: obtaining a destination point of the target object according to a travel path of the target object; acquiring a travel point in the travel path, and determining an object passing through the travel point as an object to be confirmed; selecting an object to be confirmed having the same destination point as the object as an object data set; according to the object data set, determining a first concomitant object set by an association analysis algorithm, thereby solving the technical problem of low accuracy of current concomitant vehicle discovery methods in the priorart.

The invention discloses a method, device and terminal for detecting game plug-in. The method comprises the following steps of: acquiring data to be detected of an object to be detected; detecting datato be detected based on a plurality of supervised learning models to obtain a plurality of classification results corresponding to the plurality of supervised learning models, wherein the plurality of supervised learning models are used for training data of a plurality of objects to obtain a plurality of classification results; processing a plurality of classification results from a preset dimension to obtain a degree of suspicion corresponding to the preset dimension; and judging whether the object to be detected is an abnormal object or not according to the suspicion degree corresponding tothe preset dimension, wherein the abnormal object has a plug-in behavior. The method, device and terminal for detecting game plug-in solves the technical problem of low accuracy of detecting the gameplug-in in the prior art.

The invention discloses a verbal skill recommendation method and device and a storage medium. The verbal skill recommendation method comprises the following steps: acquiring audio stream data in a first session, the audio stream data carrying request audio sent by a first user terminal to a second user terminal; converting the request audio into a session text in real time; calling a target verbalskill library, searching a target verbal skill set matched with the session text, the verbal skill for recommendation stored in the target verbal skill library being output by a verbal skill learningmodel, and the verbal skill learning model performing machine learning by using a quality inspection result output by a verbal skill quality inspection engine; and recommending the target verbal skill set to the second user terminal, so that the second user terminal determines a target verbal skill used for being pushed to the first user terminal through the first session from the target verbal skill set. The technical problem that a verbal skill recommendation method provided by related technologies is low in accuracy is solved.

The invention discloses a method for acquiring road passing time and electronic equipment. The method comprises the following steps: acquiring an intersection grade of an intersection to be passed when a vehicle passes through a target road, and acquiring a steering angle and a hindering edge grade of the intersection to be passed, wherein the hindering edge grade is a grade of passing through a road corresponding to a traffic light for waiting of the intersection to be passed according to the steering angle; acquiring needed intersection waiting time for passing through the intersection to be passed according to the intersection grade, the steering angle and the hindering edge grade; and acquiring road passing time for passing through the target road according to the intersection waiting time and a road driving speed of the target road. According to the technical scheme, the relatively accurate intersection waiting time can be acquired according to the hindering edge grade so that the road passing time is acquired; and the technical problem in the prior art that the accuracy of acquiring the road passing time is relatively low is solved, and the accuracy of acquiring the road passing time is improved.

The present invention provides a network abnormal behavior detection method and device. The method comprises: obtaining the network operation behavior of a client, combining the early-stage network operation behavior related to the network operation behavior and the network operation behavior to form a network operation behavior combination; according to the network operation behavior sequence database and the network operation behavior combination, obtaining the network operation behavior sequence related to the network operation behavior; determining whether the network operation behavior is an abnormal network operation behavior sequence or not according to the current period changing condition of the execution quantity of the network operation behavior sequence and the preset period changing condition of the execution quantity of the network operation behavior sequence; if the network operation behavior sequence is the abnormal network operation behavior sequence, rejecting the execution of the network operation behavior; and if the network operation behavior sequence is a normal network operation behavior sequence, executing the network operation behavior. The detection accuracy is high, and the network abnormal behavior detection method and device can better avoid checking mistake or checking leakage.

The invention relates to a time series data smoothing method and device. The method comprises the steps: obtaining target time series data and an associated data sequence corresponding to the target time series data, and enabling the associated data sequence to comprise the target time series data and associated data having a time series relation with the target time series data; determining a smooth weight corresponding to each part of data in the associated data sequence, the smooth weight being used for indicating the degree of difference between each part of data and the target time seriesdata and the relevancy of the distance between each part of data and the target time series data to the target time series data; and determining smooth data corresponding to the target time series data according to the smooth weight and the target time series data. According to the invention, the technical problem of low accuracy of the time series data is solved.

The invention relates to the technical field of aircraft composite material interlayer tensile stress design, and provides a method for determining an allowable value of an interlayer tensile stress design of an airplane composite material, which comprises: 1, determining influence parameters including a B reference value coefficient, an environment, a laying proportion, a thickness, a corner radius joint thickness, a damage and a thermal cycle influence coefficient; 2, determining an interlayer tensile stress design allowable value calculation formula based on the B reference value; 3, determining a test matrix; 4, determining a test environment standard, an environmental influence test moisture absorption saturation determination standard, a thermal cycle influence test cycle index and temperature spectrum, and a test piece damage determination standard; 5, performing a test by referring to an ASTM D6415 test standard to obtain an interlayer tensile stress average value and a B reference value initial value; 6, calculating a B reference value and each influence parameter; and 7, calculating an interlayer tensile stress design allowable value. The method is high in accuracy, highin universality, low in test cost and high in test efficiency.

The embodiment of the invention provides an off-network prediction method and device, relates to the technical field of communication, and solves the technical problem that an existing off-network prediction method is relatively low in accuracy. The method comprises the following steps that: an off-network prediction device firstly acquires service information of a terminal at a first moment, andthen predicts off-network information of the terminal according to the service information of the terminal and a pre-constructed decision tree; wherein the service information comprises: service information, wherein the first duration is between the opening moment of the service and the first moment; wherein the second duration is between the termination moment of the service and the first moment,whether the service comprises at least one of an internet protocol television (IPTV) service, the fault frequency of the service in a first preset time period, the total flow of the service in a second preset time period, a third duration in which the transmission flow of the service meets a preset flow threshold, the bandwidth of the service, the regional type of the service or the flow type ofthe service; wherein the off-network information is used for indicating whether the terminal is off-network at the termination moment of the service.

The invention discloses a health information chaining method and device, a storage medium and an electronic device. The method comprises the following steps: carrying out identity verification on a target object through a first biological characteristic, wherein the first biological characteristic is obtained by collecting the target object, and the target object is an object with health information to be obtained; collecting health information of the target object under the condition that the target object passes the identity verification; and uploading the health information of the target object to a block chain. The technical problem of low accuracy of recorded health information in related technologies is solved.

The invention provides a congestion risk monitoring method which includes the steps of obtaining regional parameter information and historical crowd density of a monitored region, counting the statistics user quantity of users in the monitored region, correcting the statistics user quantity according to the regional parameter information to obtain the corrected user quantity, determining the effective area of the monitored region according to the regional parameter information of the monitored region, and determining the congestion risk level of the monitored region according to the corrected user quantity, the effective area of the monitored region, and the historical crowd density of the monitored region. The invention further provides a congestion risk monitoring device. The congestion risk level in a monitored area is monitored according to the regional parameter information of the monitored region, user characteristics and the historical crowd density of the monitored region, and the monitoring accuracy is high.

The invention discloses an optical cable anomaly detection method and device. The method comprises the following steps: sending a detection signal with a preset wavelength to a to-be-detected target optical cable; receiving a feedback signal fed back by the detection signal, and determining a power curve of the feedback signal; performing wavelet transformation on the power curve to obtain an attenuation curve; and according to the attenuation curve, determining the abnormal event of the target optical cable and the position thereof. According to the invention, the technical problem of low accuracy of optical cable fault detection in the prior art is solved.

The invention discloses a knowledge graph embedding training method and a related device. The method comprises the steps of obtaining network topology information in a knowledge graph structure; calculating a first similarity distance between different entity nodes in the knowledge graph according to a similarity calculation method and the network topology information; calculating a second similarity distance between the positive sample and the negative sample based on the first similarity distance and entity nodes included in the positive sample and the negative sample in the knowledge graph; calculating a comprehensive weight corresponding to each negative sample according to the second similarity distance corresponding to each negative sample; and according to the positive sample, the negative sample, each negative sample and the corresponding comprehensive weight, carrying out model training of a corresponding type to obtain knowledge graph embedding representation. The technical problem that in the existing training process of knowledge graph embedding, different negative samples are regarded as the same kernel, and consequently the result accuracy of embedding representation is possibly low is solved.

The invention provides a CAD prediction model building method and device and electronic equipment, and relates to the technical field of modeling, and the method comprises the steps: obtaining genotype data and physical condition data of a sample object; calculating by utilizing the genotype data to obtain a CAD multi-gene risk value; and establishing a CAD prediction model based on the CAD multi-gene risk value and the physical condition data. The technical problem of relatively low accuracy of data obtained by performing CAD prediction at present in the prior art is solved.

The invention discloses an object recognition method and device, a storage medium and electronic equipment. The method comprises the steps of obtaining a combined knowledge graph corresponding to a to-be-identified interest point object combination, converting the combined knowledge graph into a combined knowledge tree, and allocating a position code to each tree node in the combined knowledge tree; inputting the combined knowledge tree and the position code into a text recognition network to obtain an object text feature corresponding to the interest point object combination; inputting a combined feature formed by splicing the object text feature and the object space feature into an object recognition network; and under the condition that the identification result output by the object identification network indicates that the similarity between the objects in the interest point object combination is greater than or equal to a target threshold, identifying the objects in the interest point object combination as the same object. The invention can be used for map associated scenes. The technical problem of relatively low accuracy of a result of identifying and distinguishing similarPOI entity objects is solved.

The invention discloses a data exception identification method and device, a storage medium and electronic equipment. The method comprises the steps of obtaining a data identification request; in response to the data identification request, obtaining attribute statistical characteristics and behavior sequence data corresponding to the log data, wherein the attribute statistical characteristics include characteristics obtained by performing statistics on parameter changes of a plurality of attribute parameters associated with a target account in the target time period, and behavior sequence data corresponding to the log data, and the behavior sequence data comprises time sequence data of behaviors executed by a virtual object controlled by the target account in a target time period; inputting the attribute statistical characteristics into a conversion model to obtain attribute combination characteristics output by the conversion model; and inputting the attribute combination features, the attribute statistical characteristics and the behavior sequence data into an identification model to obtain a data identification result output by the identification model. According to the invention, the technical problem of low accuracy of data exception identification is solved.

The invention discloses a media content pushing method and device, a storage medium and an electronic device. The media content pushing method comprises the steps: acquiring a push request of a firstterminal, wherein the push request is used for requesting to push media content to the first terminal; obtaining a first log of the first terminal, a second log of the second terminal and a relation vector between the second terminal and the first terminal, wherein media content browsed by the first account on the first terminal is recorded in the first log, and media content browsed by the secondaccount on the second terminal is recorded in the second log, and the relation vector is used for representing the relation between applications installed on the first terminal and the second terminal; selecting target media content from the media content library based on the first log, the second log and the relationship vector; and in response to the push request, pushing the target media content to the first terminal. According to the media content pushing method and device, the technical problem of relatively low accuracy of media content recommended to the user in the related art is solved.

The invention discloses a method and a device for determining the degree of acetabular bone defect. The method comprises the following steps: obtaining image data of acetabular part of a target object; converting image data into three-dimensional data; dividing the three-dimensional data according to a predetermined mode to obtain the divided three-dimensional data; analyzing the divided three-dimensional data and get the analysis result, and determining the degree of acetabular bone loss corresponding to the acetabular site according to the results of the analysis. The invention solves the technical problem in the related art that the accuracy of the evaluation result obtained by the evaluation method for the acetabular bone defect in the hip joint revision operation is low.

The invention provides an application program identification method and device. The method comprises the steps that icons of to-be-identified application programs are acquired, feature extraction is carried out on the icons of the to-be-identified application programs to obtain icon features of the to-be-identified application programs, according to the icon features of the to-be-identified application programs and reliable application programs, icon similarity between the to-be-identified application programs and the reliable application programs is determined, and according to the icon similarity, it is identified whether or not the to-be-identified application programs are abnormal. According to the icon similarity between the to-be-identified application programs and the reliable application programs, it is identified whether or not the to-be-identified application programs are abnormal, and the problems that application program names are adopted to identify application programs inthe related art, so that the abnormal application programs easily bypass identification, and the identification accuracy is lower are solved.