49 results about "Preference index" patented technology

In an embodiment, the disclosure provides a light source comprising at least one solid state light emitter. The light source, in operation, emits substantially white light having a Lighting Preference Index (LPI) of at least about 105, and this emission from the light source comprises a UV-violet flux of at least about 1%. Use of the lamps, light sources, and methods of the present disclosure may afford the ability to display linens and clothing under energy-efficient LED-based illumination, and may impart an effect to (especially white) clothing, that makes them look cleaner than under illumination by prior art LED lamps.

A management module registers a request to convert code from a first code set to a second code set, identifies a code set converter (CSC), determines whether a most recently used CSC is the identified CSC and, in response to determining that the most recently used CSC is not the identified CSC, locates a user-preferred CSC pool comprising a subset of locally stored CSCs each corresponding to a preference index. The management module, in response to determining that the identified CSC is located within the user-preferred CSC pool, searches the user-preferred CSC pool for the identified CSC, loads the identified CSC, initiates the identified CSC that converts the code from the first code set to the second code set, modifies usage data for the identified CSC, and orders the user-preferred CSC pool based on a preference index for each CSC stored in the CSC pool.

The invention provides a testing device and a testing method for testing environmental preference behaviors of fishes. The device is characterized in that water inlet pipes with a flow regulator and water outlet pipes with a flow regulator are respectively installed at the two ends of a water trough, one ends of the water inlet pipes are communicated with a pressure-bearing water tank, flow conditioning meshes are respectively installed at the communicating ports of the water inlet pipes which are communicated with the water channel, upper flow guide baffles are installed between the water inlet pipes and the water outlet pipes to form flow guide channels, isolating nets are installed at the front ends of the water inlet pipes and the water outlet pipes, and the pressure-bearing water tank is connected with a variable-frequency water pump and a variable-frequency regulator. The method comprises the steps of 1) preparation before testing; 2) testing and correction of the testing device; 3) testing of preference behaviors of fishes; 4) testing data acquisition and preference index calculation; 5) maintenance and cleaning of the device. The testing device and the testing method for testing the environmental preference behaviors of the fishes can realize researches on the preference behaviors under various environmental factors such as flowing speed, river bed materials and illumination according to different environmental factors which are distributed in an experimental area.

An embodiment of the invention provides a method and a device for adjusting recommendation lists. The method includes acquiring application download history records of users, computing proportions of game application and non-game application to total downloaded application of the users within a set time on the basis of the application download history records so as to acquire game application preference indexes and non-game application preference indexes, and generating redundancy recommendation lists for the users; adjusting the redundancy recommendation lists on the basis of the computed game application preference indexes and the computed non-game application preference indexes to form final recommendation lists; displaying the final recommendation lists for the users. Distribution of game application and non-game application in the final recommendation lists keeps corresponding to the game application preference indexes and the non-game application preference indexes of the users. The method and the device have the advantage that personalized recommendation can be carried out by the aid of the recommendation lists according to difference of interest and preference of the different users.

The invention discloses an optimal setting method for three primary colors of display equipment with a large color gamut. The method comprises the following steps: firstly, a certain number of sampling points of the three primary colors namely red, green and blue are selected in a CIE 1976UCS uniform chromaticity scale diagram to constitute imaginary color gamuts with different settings; through chromaticity characterization models built for the working states of the imaginary color gamuts, CIELAB color differences of three typical memory colors namely skin color, grass green and skyblue and CIELAB reference values thereof are calculated, so that the color preference index of each imaginary color gamut is obtained, therefore, the imaginary color gamut with the highest color preference index is selected, and the chromaticity coordinates of the three primary colors are used as optimal settings. According to the invention, the problems of color distortion when a television signal and a digital image based on small color gamut are transmitted to the display equipment with the large color gamut is solved, the operation is simple, visual evaluation is not required and the practicability is high.

A K-nearest neighbor and multi-class merge based density peak clustering method and image segmentation system are disclosed, which use density calculation mode to describe sample distribution and adopt new evaluation index to obtain clustering center. Iterative allocation strategy is designed to classify the remaining points accurately. A local class merge method is given to prevent class splitting that will contain multiple density peak points. The invention describes the distribution of each data point through the density measurement method, constructs a preference index based on the densityand the distance to evaluate the clustering center, and utilizes the iterative distribution strategy to allocate the remaining points to improve the clustering accuracy. After completing local clustering, multi-class merging strategy is used to complete the local class merging operation. The results of numerical experiments show that it has good applicability in 22 test data sets and real data.

The invention provides a sequence recommendation method based on a self-attention auto-encoder, and the method comprises the steps: obtaining a user commodity sequence and a scoring matrix, and carrying out the preprocessing of the commodity sequence; training the commodity sequence by using a self-attention model, and predicting a relevance score of the user and the commodity; reconstructing thescoring matrix by using an auto-encoder, and calculating a user preference index; and in combination with the relevance score of the user and the commodity and the user preference index, obtaining a high-score commodity to preferentially recommend to the user. According to the sequence recommendation method based on the self-attention auto-encoder, the article browsing sequence of the user is converted into a low-dimensional dense vector by using a word embedding method; the position codes are combined and input into the self-attention model, then the self-encoder is used for fitting and reconstructing the scoring matrix, the user preference index is calculated, finally, the final prediction score is obtained, recommendation is made for the user, factors such as long-term and short-term preferences of the user are considered at the same time, and the recommendation precision is effectively improved.

The present invention discloses an application recommendation method, a client end, a server and an application recommendation system. The application recommendation method comprises the following steps that: the application preference indexes of a target user for candidate applications and the application data of the candidate applications are obtained; the prospective benefit coefficients of the candidate applications are obtained according to the application preference indexes of the target user for the candidate applications and the application data of the candidate applications; and a predetermined number of recommended applications are selected from the plurality of candidate applications based on the obtained prospective benefit coefficients of the plurality of candidate applications, so that an application recommendation list can be generated and is displayed for the target user. With the application recommendation method, the client end, the server and the application recommendation system of the present invention adopted, an application platform providing application recommendation service can satisfy the requirements of user experience and maximize revenues.

The invention discloses a method of determining user risk preference, an information recommendation method and devices, so as to improve the accuracy of the determined user risk preference. The method of determining the user risk preference comprises steps: user data of a user generated in risk-involved service are acquired; according to the user data, the characteristic value of the user under each set variable in multiple set variables is determined, wherein the multiple set variables at least comprise a set variable influencing the risk preference degree of the user; and the characteristic value of the user under each set variable is inputted to a risk preference model, and the output of therisk preference model is determined to be a user risk preference index, and the risk preference index is used for measuring the risk preference degree of the user.

Aspects of the present disclosure are directed to a composite light source which includes at least one blue light source having peak wavelength in the range of about 400 nm to about 460 nm; at least one yellow-green garnet phosphor; and at least one narrow-band red-emitting down-converter. Such composite light source may have a Lighting Preference Index (LPI) of at least 120. In other aspects the disclosure is directed to composite light source comprising at least one blue light source having peak wavelength in the range of about 400 nm to about 460 nm; at least one yellow-green garnet phosphor; and at least one broad red down-converter. In this latter aspect the composite light source may have a Lighting Preference Index (LPI) of at least 120. Numerous other aspects are provided.

The present invention provides a control method for realizing accurate advertisement push based on user preference information. The control method specifically includes: a, determining a user's preference index in one or more categories based on user history browsing information and current browsing information of the user; b, comparing the preference index of one or more categories of the user with one or more advertisement information in one or more categories to determine to-be-advertised advertisement information associated with the preference index in the category with the highest similarity; and c, pushing the to-be-advertised advertisement information. The method can save the time for the user to browse the advertisement, the user is relieved of the trouble of receiving excessive unrelated advertisements, and the method has high commercial value.

The invention provides a broadcasting and television program recommendation system and method. The method includes the steps: inputting an object to be recommended, an audience preference index and an audience time period; storing program playing information as a program data source; storing audience information as an audience data source; calling the program data source and the audience data source and calculating the audience preference index of each user to form an audience preference matrix; calculating a connection relation value between the users according to the audience preference matrix, taking each user as a node and building a path between two users to form a connection network when the connection relation value between the users is larger than a set threshold value; determining credibility between the users according to the node and the path; determining the recommendation level of different users for different programs according to credibility and audience preference; selecting a set number of users according to the sequence of credibility from high to low, selecting a set number of programs according to the sequence of the recommendation level of the set number of users from high to low, and generating a program list of each user.

The invention provides a method for screening a unidirectional set-top box advertisement. The method comprises the following steps: setting that a preference index of a user to each program is the same when a set-top box leaves the factory; the set-top box counting the program watching conditions of the user at intervals of a certain unit time when the user watches the programs; if the time of watching one program of the user exceeds the specific time, accumulating the preference index corresponding to the program and storing the preference index; summing the preference index of each program in every classification according to a classification label of each program and the corresponding preference index value to obtain and store the total preference index corresponding to each program type; the set-top box receiving advertisements containing a main type label at the current moment, and querying the total preference index corresponding to the program type matched with the advertisement main type label, thereby only pushing the user interested advertisement and filtering the uninterested advertisement. Through the adoption of the method and the system disclosed by the invention, the advertisement is screened according to the user preference, the hit rate of the advertisement is improved, and the method and system are suitable for the set-top boxes in various types.

The invention discloses an individual gourmet recommending method based on the position. The distance similarity between a user and a merchant is calculated; the similarity between a user ui and a user uj is calculated; the predicted preference index of a target user to a restaurant is calculated according to the similarity between the user ui and the user uj; the similarity between an item i and an item j is calculated; the recommending preference index to a user u based on the item t is calculated; a recommending result is given according to the distance similarity between the user and the merchant, the predicted preference index of the target user to the restaurant, and the recommending preference index to the user u based on the item t. The individual gourmet recommending method has the advantages that the individual restaurant meeting the user requirement is provided for the user according to the current user position, the user grade for the merchant and the attribute of the user and the merchant, and the gourmet recommending efficiency and accuracy are improved. Meanwhile, the accurate recommending result can be converted into the consuming behavior, and the user satisfaction and merchant benefits are improved.

The invention discloses a mobile user tag based personalized food recommendation method. The method comprises the following steps that a user-restaurant location preference index is got; a restaurant-tag preference index is got, and a restaurant-tag preference vector is obtained; a user-tag preference index is got, and a user-tag preference vector is obtained; according to the restaurant-tag preference vector and the user-tag preference vector, a user-restaurant preference index is got; according to the user-restaurant preference index, the similarity of users is got; according to the similarity of the users, a predicted preference index of a target user to a target restaurant is got; according to the user-restaurant location preference index and the predicted preference index of the target user to the target restaurant, a mean absolute deviation MAEua of the target user ua is got, and therefore recommending is carried out. The mobile user tag based personalized food recommendation method has the advantages that by combining the user, the target of a merchant and the position, the personalized food recommending is carried out, the recommending quality is improved, and the recommending speed is increased.

In an embodiment, the disclosure provides a light source comprising at least one solid state light emitter. The light source, in operation, emits substantially white light having a Lighting Preference Index (LPI) of at least about 105, and this emission from the light source comprises a UV-violet flux of at least about 1%. Use of the lamps, light sources, and methods of the present disclosure may afford the ability to display linens and clothing under energy-efficient LED-based illumination, and may impart an effect to (especially white) clothing, that makes them look cleaner than under illumination by prior art LED lamps.

The invention provides a recommending method based on high-order user preferences. The method comprises the following steps of: using a user-object grading data set to calculate a high-order preference value of a user on an object, and building a user-object-object paired preference data set; initializing a model, and randomly selecting values from normal distribution to initialize the model; selecting relevant data from the grading data set and the paired preference data set; calculating errors and the user preference similarity to form an optimization criterion; calculating the preference index gradient, and updating the model; and repeating the steps till the model parameter convergence. The method provided by the invention consists of an OPTRP optimization criterion and an LearnRP learning algorithm; through the learning, an existing CF model can be generated; a new model can also be generated; a highly deviated grading data set can be effectively processed, so that the prediction precision and the recommending precision of a recommending system are improved; and important application values are realized in real scenes.

The invention provides a set-top box program skipping and sorting method. The method comprises the following steps: setting that initial values of a preference index and a skipping index of a user to each program are the same when a set-top box leaves the factory; the set-top box counting the program watching conditions of the user at intervals of a certain unit time when the user watches the programs; the set-top box judging whether the corresponding program needs to skip according to the counted skipping index of each program; if the skipping index of one program exceeds a skipping threshold value, the set-top box automatically skipping the program when the user up or down switches the channels; the set-top box sorting the whole program list according to a counted accumulation value of the preference index of each program so as to automatically switch according to the sorted program list order when the user up or down switches the channels. The method disclosed by the invention has the advantages that the corresponding program skipping and whole program list sorting can be automatically performed according to the preference degrees of the user to the programs by counting the program watching conditions of the user through the set-top box, and the set-top box is more convenient and intelligent.

According to some embodiments, a composite light source includes at least one blue light source having a peak wavelength in the range of about 400 nanometer (nm) to about 460 nm; at least one LAG phosphor; at least one narrow red down-converter; and wherein the composite light source has a Lighting Preference Index (LPI) of at least 120. Numerous other aspects are provided.

The invention provides a network access switching method and device. Whether the current network resource meets the need of a user is periodically determined. If the current network resource does not meet the need of the user, access network switching is triggered. The access evaluation index of each alternative network and the preference index of the user to each alternative network are calculated, and the switching value index of each alternative network is calculated. The switching value index is used as the basis for selecting a target switching access network. According to the invention, the algorithm is simple, and takes into account the user preferences, the performance of each current access network, the performance of each access device and other factors; the selected access network not only meets the need of the user, but also matches the current network performance and the device performance; and the rationality of the selection result of the access network is improved.

The invention relates to big data, and discloses a product recommendation method and device based on big data, equipment and a storage medium. The big data-based product recommendation method comprises the steps of reading user account information of a to-be-recommended product, the user account information comprising a plurality of user behavior indexes; inputting the user account information into a preset preference index evaluation model, and counting scores of the plurality of user behavior indexes through the preference index evaluation model to obtain a first preference index corresponding to the satisfaction degree of the user account for each type of products; screening products corresponding to the user account based on the first preference index to obtain a corresponding productrecommendation scheme; and packaging the product recommendation scheme and pushing the product recommendation scheme to a corresponding user account. In addition, the invention also relates to a blockchain technology, and the user account information can be stored in the blockchain. By analyzing all the behavior indexes of the user, the accuracy of generating the product recommendation scheme forthe user is higher.

The invention discloses a wireless heterogeneous network load balancing method based on feedback calculation. The method comprises the following steps: calculating an effective value of a candidate network according to a user preference index, an application requirement index and a terminal index; calculating a feedback factor; and calculating an evaluation value of the candidate network by using the feedback factor and the effective value of the candidate network; and selecting a target network from the candidate network by using the evaluation value, and selecting to switch the current network or retain the current network according to the result of the target network. By adoption of the method, the load balancing effect.

In an embodiment, the disclosure provides a light source comprising at least one solid state light emitter. The light source, in operation, emits substantially white light having a Lighting Preference Index (LPI) of at least about 105, and this emission from the light source comprises a UV-violet flux of at least about 1%. Use of the lamps, light sources, and methods of the present disclosure may afford the ability to display linens and clothing under energy-efficient LED-based illumination, and may impart an effect to (especially white) clothing, that makes them look cleaner than under illumination by prior art LED lamps.

The invention relates to the big data analysis, and provides a campus canteen management method, an electronic device and a computer-readable storage medium. The information of the ordering end is transmitted to the serving end through a server end. The ordering end displays a campus map and the cuisine information of each canteen. A point menu is generated according to the cuisine information selected at the ordering end. The order menu generated by the order end is sent to the supply end of the corresponding canteen identifier through the server end; After the menu is taken away from the ordering end, the order completion signal generated by the vegetable supply end is sent to the ordering end through the server end; the preference information of the order end to the dishes is obtained;the preference index of the order side to the dishes preference degree is constructed. The dishes preference matrix is constructed according to the preference information and preference index of eachorder; the ordering end are clustered and grouped according to the dishes preference matrix to obtain a set number of user groups. The method, the device and the medium can view the canteen position,the canteen cuisine and the price, and can cluster and classify according to the user preference.

The invention relates to a doctor-patient matching method based on a Bayesian network, and the method comprises the steps: S1, collecting electronic medical record data, determining disease and symptom nodes and values thereof, and enabling the disease and symptom nodes to serve as training set data; S2, constructing a disease-disease / disease-symptom self-interaction matrix, and taking the disease-disease / disease-symptom self-interaction matrix as a constraint of a Bayesian network; S3, constructing a Bayesian network model, and performing structure learning and parameter learning; S4, inputting the disease into the Bayesian network pre-diagnosis model by the patient to obtain all possible disease combinations for calculating the complication of the main disease and the complication; S5, calculating a matching index of doctors and patients; S6, constructing a doctor recommendation model based on the matching index of the doctors and the patients; and S7, obtaining the optimal allocation of the patients and the doctors according to the preference index of the patient. According to the invention, doctor-patient matching is carried out by combining the pre-diagnosis result, the doctorspecialty, the doctor workload and the patient preference, and the defect that the current doctor-patient matching technology is not accurate enough is solved.

The application discloses a user business preference assessing method, a user investment risk preference assessing method and a user business preference assessing device. The method comprises the following steps: characteristic vectors that are generated when business operation is performed by a user are obtained; composing elements of the characteristic vectors comprise at least one business preference index value which is generated when the business operation is performed by the user; the characteristic vectors are processed according to preset rules, a degree of preference of the user toward the business operation is obtained, preference of the user toward business can be assessed according to the degree of preference, and the preset rules are obtained via training through the characteristic vectors that are generated when the business operation is performed by the user and the degree of preference of the user toward the business. According to the user business preference assessing method, the user investment risk preference assessing method and the user business preference assessing device disclosed in an embodiment of the application, the characteristic vectors that are generated when the business operation is performed by the user are obtained, and therefore the degree of preference of the user toward the business can be obtained; the degree of preference can objectively reflect the reference of the user toward the business.

The invention provides a set-top box random switching construction method based on user preferences. The method comprises the following steps: 1, initially setting equal preference indexes of a user specific to each channel, and setting unit time T; 2, counting the number of channels watched by the user according to the set unit time T; 3, reading the watching statistical data of the each watched channel, and constructing a random standby channel set; 4, generating a random number from the constructed frequency set, wherein a channel number corresponding to the random number indicates a randomly-switched channel. Through adoption of the method, channels can be generated randomly according to the watching habits and preferences of a user, and favorite channels of the user can be selected easily at a high hit rate.

The invention belongs to the technical field of computers, and particularly relates to a workflow management system testing method and device, a computer readable storage medium and terminal equipment. The method comprises the following steps: extracting each test order in a test order library, and determining a reference circulation path corresponding to each test order; obtaining historical test records obtained by testing the historical version of the workflow management system at each time from a database; according to this, calculating the deviation degree of each test of each test order is calculated, and the priority index of each test order is calculated; sorting the test orders in the test order library according to the sequence of the preference indexes from high to low; and according to a preset test time limit, selecting a front specified number of test orders in a sequence from the test order library to construct a test set of a current test, and testing a current updated version of the workflow management system by using the test set to ensure that as many as possible system defects are found within a limited time.

The present invention proposes a calculation method, system, computer equipment and computer-readable storage medium of the salient features of the target group, wherein the calculation method of the salient features of the target group includes: acquiring multiple dimensional features of the target group, wherein each dimension Features include multiple category features; calculate the correlation between the target group and each dimensional feature, and obtain multiple correlation coefficients; filter multiple dimensional features according to the correlation coefficients, and obtain multiple first-type dimensional features; calculate The group preference index of each category feature of the first type of dimension feature; compare the size relationship between the group preference index and the first preset threshold, and use the category feature corresponding to the group preference index greater than the first preset threshold as the target group salient features. The present invention can greatly shorten the calculation time of the TGI index and does not require manual intervention, greatly improves the efficiency and lowers the business threshold.