386 results about "Score matrix" patented technology

The invention discloses a collaborative filtering recommendation method for introducing program popularity weighting, which is characterized in that on the interface of an IPTV program, a visual menu for users to give marks is provided and a program recommendation list is made for target users according to user watching time, conduct operation and program marking data sent by a terminal set-top box. The invention comprises the detailed steps of: collecting the behavior characteristic information of users, working out a 'user-item' scoring matrix A(m, n), calculating popularity weight value, calculating similarity degree and sorting, making forecast score for the target users and sorting, and working out the recommendation list for the target users. Compared with the prior art, the method disclosed by the invention is more in accordance with objective reality, improves the quality of collaborative filtering and the precision degree of recommendation, initiatively cuts own the programs according to user preferences and behavior characteristics, carries out personalized recommendation to the programs which the users like and realizes the purpose that 'watch the program you like whenever you want'.

The invention discloses an internet information product recommending method based on matrix decomposition. The method comprises the following steps of: 1) obtaining the user scoring record to the information product; 2) obtaining the social relationship record between internet users; 3) respectively building a scoring matrix and a social matrix according to the types of a target user and a target product; 4) learning a user feature vector and a product feature vector through a matrix decomposition technology; 5) calculating the scores of different products scored by the target user according to the feature vectors, so as to recommending the favorite products of the user according to the scores. In the method, analysis on user social relationship is introduced, and personalized product recommendation is provided for the target user based on the production type information. The calculation is simple and quick, and the method has better expandability and adaptability, so that the method is suitable for highly dynamic and immense amount of product-oriented recommendation for the internet users.

Disclosed herein are a method and system for training a big data machine to defend, retrieve log lines belonging to log line parameters of a system's data source and from incoming data traffic, compute features from the log lines, apply an adaptive rules model with identified threat labels produce a features matrix, identify statistical outliers from execution of statistical outlier detection methods, and may generate an outlier scores matrix. Embodiments may combine a top scores model and a probability model to create a single top scores vector. The single top scores vector and the adaptive rules model may be displayed on a GUI for labeling of malicious or non-malicious scores. Labeled output may be transformed into a labeled features matrix to create a supervised learning module for detecting new threats in real time and reducing the time elapsed between threat detection of the enterprise or e-commerce system.

The invention discloses a cold-chain logistic stowage intelligent recommendation method based on spectral clustering. Scores of users for a stowage line are conveyed through a cold chain for cold-chain logistic stowage intelligent recommending, a score matrix is built, the Euclidean distance is used for calculating the user similarity, a degree matrix is used for calculating a Laplacian matrix, feature vectors are obtained by calculating feature values of the orderly Laplacian matrix, a K-means algorithm is used for clustering the feature values to obtain a user group with the similar interesting stowage line, and a stowage line is recommended inside the user group with the similar interesting stowage line, so that cold-chain logistic stowage intelligent recommending is achieved, the cold-chain logistic vehicle non-load ratio is lowered, and the profit rate of cold-chain logistic transport vehicles is increased.

Methods and systems for mapping an event type to an activity in a business process model are disclosed. In accordance with one such method, the event type and the activity are tokenized by determining event tokens for event type labels in the event type and determining activity tokens for activity labels in the activity. In addition, a score matrix is generated for pairs of the event tokens and the activity tokens indicating a degree of similarity between the event token and the activity token in each of the pairs. The method also includes determining whether the event type and the activity are correlated by determining scores of the pairs of event tokens and activity tokens that are ranked highest in said score matrix. Further, a mapping report indicating whether the event type and the activity are correlated in the business process model is output.

A method and apparatus for performing discriminative training of, for example, call routing training data (or, alternatively, other classification training data) which improves the subsequent classification of a user's natural language based requests. An initial scoring matrix is generated based on the training data and then the scoring matrix is adjusted so as to improve the discrimination between competing classes (e.g., destinations). In accordance with one illustrative embodiment of the present invention a Generalized Probabilistic Descent (GPD) algorithm may be advantageously employed to provide the improved discrimination. More specifically, the present invention provides a method and apparatus comprising steps or means for generating an initial scoring matrix comprising a numerical value for each of a set of n classes in association with each of a set of m features, the initial scoring matrix based on a set of training data and, for each element of said set of training data, based on a subset of said features which are comprised in the natural language text of said element of said set of training data and on one of said classes which has been identified therefor; and based on the initial scoring matrix and the set of training data, generating a discriminatively trained scoring matrix for use by said classification system by adjusting one or more of said numerical values such that a greater degree of discrimination exists between competing ones of said classes when said classification requests are performed, thereby resulting in a reduced classification error rate.

The invention relates to a commodity recommendation method and device based on an electronic business platform and a server. The commodity recommendation method and device based on the electronic business platform and the server are used for improving the accuracy of individual commodity recommendation. The method comprises the steps that adhesive user groups are screened; user interestingness is acquired; a neighbor set closest to a target user is generated; a commodity recommendation list is generated. According to the technical scheme, the obtained adhesive user groups can be more accurate and efficient; due to the fact that only the interestingness of each user in the adhesive user groups is recorded in an obtained scoring matrix, an evaluation matrix can be filled, and the density of the scoring matrix is improved; the closest neighbor set between the target user and the other users in the adhesive user groups is calculated through the scoring matrix, the similar neighbors of the target user can be searched for more accurately, and then the accuracy of recommending commodities to the target user is improved.

The invention discloses a recommendation system optimization method with information of a user and an item and a context attribute integrated. According to the method, the information of the user, the item and the context attribute is integrated in a matrix decomposition model, and recommendation accuracy is improved in a personalized recommendation system. The recommendation system optimization method with the information of the user, the item and the context attribute integrated is characterized in that different influences of the information of the user, the item and the context attribute on overall scores, user interests and item scores are considered, and is applied to calculation of an original matrix decomposition model. The influences of the user, the item and the context attribute on the scores are considered at the same time, and therefore the recommendation accuracy is obviously higher than that of the rectangular decomposition model which only adopts user program two-dimensional score matrix information.

The invention relates to a collaborative filtering method for personalized recommendation fusion content and behavior. The method comprises the following steps of, (1) characteristic input, includinga project-attribute matrix representing a project content and a user behavior matrix representing user behaviors; (2) content-based project clustering for calculating the similarity of projects and clustering the projects; (3) score prediction and feature filling including carrying out score prediction on the non-scoring projects, and filling a user-project scoring matrix; (4) behavior-based userclustering including clustering users according to a project clustering result and a user-project scoring matrix; (5) score predication and project recommendation including determining the clusteringcluster where the target users are located, finding a nearest neighbor user set, performing score prediction on the non-scoring projects of the target users, and finally recommending the first N projects with the highest prediction scores to the target users. Compared with the prior art, the collaborative filtering method effectively solves the problems of data sparsity and cold start, and ensureshigh recommendation efficiency.

The invention discloses a driving behavior analyzing method based on vehicle-mounted data. The driving behavior analyzing method comprises the following steps of collecting driving data of automobiles in a real time manner; deleting invalid driving data according to bit mask; performing deletion or modification on exceptional data; performing descriptive statistic on the driving data so as to obtain statistic driving data; through an analytic hierarchy process of group decision, obtaining the weight distribution of evaluating indicators; through a method for scoring by an expert, obtaining scores during a stroke period; through a conformability principle, obtaining the duration of the stroke, the average velocity of the stroke, and the ride comfort scores of the stroke; and according to the weight distribution of the evaluating indicators and the score, obtaining a scoring matrix of the driving data. The invention further discloses a driving behavior evaluation system based on the vehicle-mounted data. According to the driving behavior analyzing method and evaluation system based on the vehicle-mounted data, through the level analysis method of the group decision, the weight of the driving data is obtained, and through the conformability principle, the score of the driving data is obtained, so that the driving behavior of a user can be accurately analyzed and evaluated.

A method of doing business is disclosed for the sale of insurance options. An individual is assigned a Health Risk Score (HRS); in a preferred embodiment, the HRS is determined by using an automated case-based method for assessing risk. The individual may then purchase an option for insurance coverage that will begin only after a predetermined period of time. At the beginning of that period, the individual's HRS is re-evaluated, and if it remains within predetermined acceptable limits, the individual may then obtain insurance coverage with premiums and benefits defined at the time of the purchase of the option. A case-based method for assessing risk is disclosed, in which various risk factors and their combinations are used to provide a scoring matrix. This scoring matrix is then used to evaluate the individual's HRS.

The invention discloses an item-based explicit and implicit feedback mixing collaborative filtering recommendation algorithm. The method comprises the following steps of obtaining the information of interest of users on every item and establishing the score matrix of every user on all the items; calculating the average score of every user, the quantity of the scoring users of every item and the average score of every item; calculating a common comment user quantity matrix; calculating the Pearson similarity and the modified cosine similarity of between any two items; calculating the similarity based on explicit feedback; calculating the cosine similarity based on implicit feedback; calculating a final similarity; obtaining the nearest neighbor set I of a current item; when providing a recommendation list to a target user u, according to the score matrix, obtaining the scored items and the unscored items of the target user u; calculating the prediction scores of the unscored items of the target user u and selecting N items with the highest scores inside the unscored items of the target user u to the user. The item-based explicit and implicit feedback mixing collaborative filtering recommendation algorithm can effectively improve the accuracy of prediction recommendation.

The invention provides for transformation of a section of a data block independently of the transformation of separate or overlapping data blocks to determine a property related to the original matrix, where each of the separate or overlapping data blocks are derived from an original data matrix. The transformation enhances parameters of a first data block over a given region of an axis of the data matrix, such as signal-to-noise, without affecting analysis of a second data block derived from the data matrix. This allows for enhancement of analysis of an analyte property, such as concentration, represented within the original data matrix. In a first embodiment of the invention, a separate decomposition and factor selection for each selected data matrix is performed with subsequent score matrix concatenization. The combined score matrix is used to generate a model that is subsequently used to estimate a property, such as concentration represented in the original data matrix. In a second embodiment, each data matrix is independently preprocessed. Demonstration of the invention is performed through glucose concentration estimation from noninvasive spectra of the body.

The invention relates to a collaborative recommendation method based on user cognition degree changes. The method comprises the steps that firstly, an RMF model is used for decomposing a user-project scoring matrix into a user feature matrix and a project feature matrix, a time attenuation function based on a psychological memory-forgetting curve is introduced, and attenuation of different degrees is given to scores given by the user at different periods of time; secondly, four aspects including user historic behaviors, the user age, occupation and sex for influences on the user cognition degree are considered, and a factor model is used for performing modeling of the four factors affecting the cognition degree on the RMF model so as to overcome defects in a traditional method. The collaborative recommendation method (CogTime_RMF) based on user cognition degree changes can construct user cognition degree changes from the multiple factors such as age, occupation, sex and historic behaviors, the cognition degree changes are integrated into the recommended scheme, and the recommending accuracy can be improved while the problem of cold boot of the user can be solved.

The embodiment of the invention discloses a television program recommending method and device for a digital television. The method comprises the following steps: collecting basic information of digital television subscribers; constructing an initial predictive scoring matrix according to the basic information of the subscribers; establishing a similar television program list according to the initial predictive scoring matrix; and acquiring the recommendation result according to the similar television program list. In the method and device provided by the embodiment of the invention, a content-based recommendation method is added on the basis of a collaborative filtering recommendation method so as to solve the problem of cold start up of a collaborative filtering recommendation system in the prior art, and the two methods are combined to construct the similar television program list, and recommendation service is provided for the subscribers by use of the similar television program list. Therefore, the method and device provided by the invention can be used for assisting the digital television subscribers to find the possible interested television programs, can ensure that a new digital television subscriber can obtain more accurate recommendation service, and can be used for solving the problem that a new television program has less possibility of being recommended by the system.

The invention relates to a video pushing method, apparatus, computer device and storage medium. The method includes acquiring a plurality of user identifiers and corresponding video identifiers and corresponding historical behavior data, calculating scoring data corresponding to each user identifier according to the historical behavior data of each user identifier, assembling a score data set, storing the scoring data set according to a matrix to obtain a scoring matrix, inputting user ID, video identification and scoring matrices into the trained video push model, decomposing the scoring matrix, according to the decomposition result, determining the user similarity between users and the video similarity between videos, according to the user similarity between users and the video similarity between corresponding videos, determining the target video identification set corresponding to each user identification, and pushing the video link corresponding to the target video identification set to the corresponding terminal corresponding to each target user identification, so as to improve the accuracy of video recommendation and enhance the user experience.

The invention relates to data processing and recommendation in a computer technology and discloses a recommendation method for searching a target user and matching a target product for an automobile industry. A corresponding automobile product is automatically recommended for the target user with a purchasing intention, so that the marketing cost of companies is reduced. The method can be concluded as follows: a, a pre-treatment phase of data mainly aims at knowing and analyzing a task of a system and reducing dimensions of original data through steps of cleaning data, integrating and simplifying and the like, so that target data used for a predication phase is generated; b, the predication phase mainly aims at finding the target user with the purchasing intention according to an association rule, and an attribute vector is obtained by knowing about preference attributes corresponding to the target user through a manner of obtaining network search recording mining of the target user or a questionnaire survey manner; finally, the similarity between users is calculated by adopting a user-attribute scoring matrix through a collaborative filtering algorithm, so that a predicated result is given based on the similarity; c, an evaluation phase aims at evaluating the predicated result. The recommendation method is applicable to recommendation of automobile products.

The invention provides a recommendation system cold start solving method based on user feedback. The method comprises the following steps: selecting data samples; constructing a time sequence sample matrix, dividing a user-commodity real score matrix into a plurality of sub matrixes according to a time sequence, simulating emergence of new users, taking sub matrixes in the top of time rank as training sub matrixes and taking other sub matrixes as test sub matrixes; and establishing a user-commodity characteristic matrix by using a latent semantic model, introducing new users into a confidence interval upper bounded UCB algorithm model and iterating and updating user characteristics and commodity characteristics. The recommendation system cold start solving method does not need extra information and is capable of rapidly screening commodities interested by the users according to limited frequency of user feedback interaction.

The invention discloses a collaborative filtering algorithm based on user and project mixing, and the algorithm comprises the steps: 1, carrying out the arrangement of a user-project scoring data set, and building a user-project scoring matrix U; 2, calculating the similarities of articles, and ordering the similarities from the big to the small; 3, generating a 'nearest neighbor N' of articles according to the similarity ordering of the articles; 4, calculating the similarities between a target user T and other users, and ordering the similarities from the big to the small; 5, generating a 'nearest neighbor N' of users according to the similarity ordering of the users. The algorithm gives consideration to the similarities of the users and the similarities of the projects, obtains a project prediction score (giving consideration to the similarities of the users and the similarities of the projects at the same time) through employing a weighting method, carries out recommendation according to the ordering of scores, can reduce the value of an MAE (mean average error), and improves the accuracy of a recommended algorithm.

According to some embodiments, account information may be received in connection with a potential insurance policy. A premium indication portal processor may receive, from a risk score model application, an account score matrix for the potential insurance policy, including grade values comparing the account information with other insured policies in a risk database, along with a benchmark premium value calibrated to a target return on equity based on the account information and information in the risk database. The account score matrix may be displayed on an underwriter device, and guide indication adjustments may be received from the underwriter device for the potential insurance policy. The premium indication portal processor may then automatically calculate an adjusted premium value calibrated to the target return on equity based at least in part on the guide indication adjustments.

The invention provides a socialized recommendation method based on a matrix decomposition and network embedding joint model. The method comprises the following steps: constructing a user-article scoring matrix and user-user social network and generating a user social corpus according to user-user social network; utilizing a user-article scoring data and a user social corpus training matrix decomposition and network embedding joint model to obtain a user feature matrix and an article feature matrix; predicting an unobserved score according to the user feature matrix and the article feature matrix; and recommending a plurality of articles with relatively high prediction score values to the corresponding users. According to the method, a matrix decomposition model and a network embedding model are seamlessly integrated by designing a unified target function; based on a unified optimization framework, bidirectional promotion and collaborative optimization between a matrix decomposition model and a network embedding model are realized, so that interested articles can be accurately recommended to a user.