Abstract: Features of multiple dimensions are extracted from information included in a URL access request. A risk score of the URL access request is obtained by providing the features to a predetermined URL attack detection model for prediction calculation, where the predetermined URL attack detection model is a machine learning model obtained through training based on the Isolation Forest machine learning algorithm. It is determined, based on the risk score, that the URL access request is a URL attack request.

Abstract: Implementations of the present specification provide abnormal sample prediction methods and apparatuses. The method includes: obtaining a sample to be tested, wherein the sample to be tested comprises feature data with a given dimension, and wherein the given dimension is a first quantity; performing dimension reduction processing on the sample to be tested by using multiple dimension reduction methods to obtain multiple processed samples; inputting the multiple processed samples to multiple corresponding processing models to obtain scores of the multiple processed samples, wherein an ith processing model Mi in the multiple processing models scores the corresponding processed sample Si based on a hypersphere Qi; determining a comprehensive score of the sample to be tested based on scores of the multiple processed samples; and classifying, based on the comprehensive score, the sample to be tested as an abnormal sample.

Abstract: Embodiments of the specification provide a URL abnormal field location method. One exemplary method comprising: obtaining a plurality of URL samples comprising a plurality of abnormal URL samples and a plurality of normal URL samples; for each of the plurality of URL samples, obtaining a plurality of feature vectors representing the plurality of fields of the URL sample; assigning a plurality of training labels to the plurality of feature vectors of each of the plurality of URL samples; obtaining, based on a classifier, a plurality of predicted labels for the plurality of feature vectors of each of the plurality of URL samples; updating the plurality of training labels based on the plurality of predicted labels; training the classifier with the plurality of updated training labels; and deploying the trained classifier to identify an abnormal field in a URL.

Abstract: The application discloses an optical network planning method for asymmetric traffic transmission over a multi-core fiber optical network and a network using the same. The method comprises: acquiring an asymmetric traffic demand over a multi-core fiber optical network to obtain a target service; establishing a corresponding route depending on the target service, and selecting cores in a multi-core fiber and allocating corresponding frequency slots in an interleaving and counter-propagating manner to each link along the route to optimize optical network planning and design. With the method provided by the application, through selecting cores in a multi-core fiber and allocating corresponding frequency slots in an interleaving and counter-propagating manner to each link along the route, the inter-core crosstalk is suppressed and network capacity efficiency is increased, thereby optimizing optical network planning and design for traffic transmission over the multi-core fiber optical network. (FIG.

Abstract: A feature extraction is performed on transaction data to obtain a user classification feature and a transaction classification feature. A first dimension feature is constructed based on the user classification feature and the transaction classification feature. A dimension reduction processing is performed on the first dimension feature to obtain a second dimension feature. A probability that the transaction data relates to a risky transaction is determined based on a decision classification of the second dimension feature, where the decision classification is based on a pre-trained deep forest network including a plurality of levels of decision tree forest sets.

Abstract: The present disclosure describes techniques for object classification using deep forest networks. One example method includes classifying a user object including features associated with the user based on a deep forest network including identifying one or more user static features, one or more user dynamic features, and one or more user association features from the features included in the user object; providing the user static features to first layers, the user dynamic features to second layers, and the user association features to third layers, the first, second, and third layers being different and each providing classification data to the next layer based at least in part on the input data and the provided user features.

Abstract: Features of multiple dimensions are extracted from information included in a URL access request. A risk score of the URL access request is obtained by providing the features to a predetermined URL attack detection model for prediction calculation, where the predetermined URL attack detection model is a machine learning model obtained through training based on the Isolation Forest machine learning algorithm. It is determined, based on the risk score, that the URL access request is a URL attack request.

Abstract: An index anomaly detection method includes: acquiring data of each of monitoring points, contained in a period of time, of a monitored index; extracting a mean value and a variance of the data of the monitoring points using a Gaussian model; calculating, according to the mean value and the variance of the data of the monitoring points, probabilities of occurrence of the data of the monitoring points, respectively; calculating, according to the respectively calculated probabilities, joint probabilities of occurrence of the data of the monitoring points contained in respective windows divided from the period of time; and detecting, according to the joint probabilities corresponding to the respective windows, whether the monitored index is abnormal.

Abstract: A wireless network access method and apparatus are provided. The method includes obtaining identification information and status information of one or more network access points, determining a target access point according to the status information, submitting the identification information of the target access point to a network access server via a mobile communications network, receiving access account information for the target access point from the network access server, and transmitting a wireless network access request including the received access account information, to the target access point.

Abstract: This specification describes techniques for detecting abnormal data in a data set. One example method includes obtaining, by a data processing platform, a to-be-validated data group including to-be-validated data corresponding to a predetermined feature; obtaining, by the data processing platform, a comparison data group including historical data associated with the to-be-validated data group, wherein the historical and the to-be-validated data are from a same data source; performing, by the data processing platform, a two-group significance test on the to-be-validated data group and the comparison data group to generate a test result; and determining, by the data processing platform, whether there is abnormal data in the to-be-validated data group based on the test result.

Abstract: By a computing platform, a classification sample set is obtained from a user operation record, where the classification sample set includes calibration samples, where each calibration sample includes a user operation sequence and a time sequence. For each calibration sample and at a convolution layer of a fraudulent transaction detection model: a first convolution processing is performed on the user operation sequence to obtain first convolution data and a second convolution processing is performed on the time sequence to obtain second convolution data; the first convolution data is combined with the second convolution data to obtain time adjustment convolution data, and the time adjustment convolution data is entered to a classifier layer of the fraudulent transaction detection model to generate a classification result; and the fraudulent transaction detection model is trained using the classification result. A fraudulent transaction is detected using the trained fraudulent transaction detection model.

Abstract: Techniques for data sharing between a data miner and a data provider are provided. A set of public parameters is downloaded from the data miner. The public parameters are data miner parameters associated with a feature set of training sample data. A set of private parameters in the data provider can be replaced with the set of public parameters. The private parameters are data provider parameters associated with the feature set of training sample data. The private parameters are updated to provide a set of update results. The private parameters are updated based on a model parameter update algorithm associated with the data provider. The update results is uploaded to the data miner.

Abstract: A feature extraction is performed on transaction data to obtain a user classification feature and a transaction classification feature. A first dimension feature is constructed based on the user classification feature and the transaction classification feature. A dimension reduction processing is performed on the first dimension feature to obtain a second dimension feature. A probability that the transaction data relates to a risky transaction is determined based on a decision classification of the second dimension feature, where the decision classification is based on a pre-trained deep forest network including a plurality of levels of decision tree forest sets.

Abstract: A feature extraction is performed on transaction data to obtain a user classification feature and a transaction classification feature. A first dimension feature is constructed based on the user classification feature and the transaction classification feature. A dimension reduction processing is performed on the first dimension feature to obtain a second dimension feature. A probability that the transaction data relates to a risky transaction is determined based on a decision classification of the second dimension feature, where the decision classification is based on a pre-trained deep forest network including a plurality of levels of decision tree forest sets.

Abstract: The present disclosure describes techniques for object classification using deep forest networks. One example method includes classifying a user object including features associated with the user based on a deep forest network including identifying one or more user static features, one or more user dynamic features, and one or more user association features from the features included in the user object; providing the user static features to first layers, the user dynamic features to second layers, and the user association features to third layers, the first, second, and third layers being different and each providing classification data to the next layer based at least in part on the input data and the provided user features.

Abstract: Techniques for identifying fraudulent transactions are described. In one example method, an operation sequence and time difference information associated with a transaction are identified by a server. A probability that the transaction is a fraudulent transaction is predicted based on a result provided by a deep learning network, where the deep learning network is trained to predict fraudulent transactions based on operation sequences and time differences associated with a plurality of transaction samples, and where the deep learning network provides the result in response to input including the operation sequence and the time difference information associated with the transaction.

Abstract: By a computing platform, a classification sample set is obtained from a user operation record, where the classification sample set includes calibration samples, where each calibration sample includes a user operation sequence and a time sequence. For each calibration sample and at a convolution layer of a fraudulent transaction detection model: a first convolution processing is performed on the user operation sequence to obtain first convolution data and a second convolution processing is performed on the time sequence to obtain second convolution data; the first convolution data is combined with the second convolution data to obtain time adjustment convolution data, and the time adjustment convolution data is entered to a classifier layer of the fraudulent transaction detection model to generate a classification result; and the fraudulent transaction detection model is trained using the classification result. A fraudulent transaction is detected using the trained fraudulent transaction detection model.

Abstract: This specification describes techniques for detecting abnormal data in a data set. One example method includes obtaining, by a data processing platform, a to-be-validated data group including to-be-validated data corresponding to a predetermined feature; obtaining, by the data processing platform, a comparison data group including historical data associated with the to-be-validated data group, wherein the historical and the to-be-validated data are from a same data source; performing, by the data processing platform, a two-group significance test on the to-be-validated data group and the comparison data group to generate a test result; and determining, by the data processing platform, whether there is abnormal data in the to-be-validated data group based on the test result.

Abstract: Provided are a door seal and a refrigeration container having the door seal. The door seal comprises a sealing air chamber portion and a latching tongue portion. The latching tongue portion is used to be connected to the door body. The latching tongue portion comprises a barb and a fixed plate for connecting the barb to the sealing air chamber portion. The sealing air chamber portion is connected to the fixed plate through a connection portion, and a gap is formed between the sealing air chamber portion and the fixed plate. Several through holes are provided on the fixed plate at a position between the barb and the connection portion. The through holes are arranged on the fixed plate without affecting the appearance. When the door body is closed, the through holes are pressed and blocked by the sealing air chamber portion without affecting the sealing.

Abstract: Techniques for data sharing between a data miner and a data provider are provided. A set of public parameters is downloaded from the data miner. The public parameters are data miner parameters associated with a feature set of training sample data. A set of private parameters in the data provider can be replaced with the set of public parameters. The private parameters are data provider parameters associated with the feature set of training sample data. The private parameters are updated to provide a set of update results. The private parameters are updated based on a model parameter update algorithm associated with the data provider. The update results is uploaded to the data miner.