67results about How to "Reasonable classification" patented technology

The embodiment of the invention provides a document classification method and device. The document classification method comprises the steps that each participle in documents to be classified is converted into a vector by training a deep neural network language model; vectors are clustered to generate a similar participle set; the documents to be classified are converted into a characteristic frequency inverse document matrix according to the characteristics of the set; the characteristic frequency inverse document matrix is converted into a hierarchical clustering tree by calculating the similarity of the vectors of any two documents to be classified; the hierarchical clustering tree is dynamically cut at different heights based on preset end conditions to obtain classified documents. According to the method, context information of participles in specific language environments is considered during document classification, and accordingly each type of documents is higher in semantic comprehension and semantic recognition. In addition, the hierarchical clustering tree is dynamically cut at different heights based on the preset end conditions, the problem that all types of documents are large in document number difference is avoided, and document classification is more reasonable.

The invention provides a method and a device for naming a photo album. The method comprises the following steps of obtaining photo information of to-be-processed photos; generating a corresponding event photo album according to the obtained photo information; obtaining the photo information of photos in the photo album, extracting the photo information content which is used as the name of the photo album from the obtained photo information according to a set name extraction rule, and generating the name of the photo album. According to the method provided by the invention, the content can be extracted according to the rule from the photo information so as to generate the name of the photo album.

The invention relates to an allocation method and an allocation system for heterogeneous convergence network resources. The method comprises a first step of judging whether a visible light network is available in a target environment covered by both the visible light network and a long term evolution (LTE) network, if yes, allowing a user in the target environment to access into the visible light network and moving onto step2, if not, allowing the user in the target environment to access into the LTE network and moving onto step3; a second step of allocating frequency resource of the visible light network for the user through an extremum method, and allocating power resource of the visible light network for the user according to attribute information of a visible light network channel; and a third step of allocating frequency resource of the LTE network for the user according to bandwidth requirement of the user, and allocating power resource of the LTE network for the user through an improved water injection method. The method and the system can allocate power resource and frequency resource for the user more reasonably in a heterogeneous convergence network environment.

The invention belongs to the technical field of monitoring of crop diseases and in particular relates to a method for determining wheat head scab levels based on a cloud model. The method comprises the following steps: extracting disease areas, constructing color characteristics, generating a gibberella saubinetii color characteristic cloud model, establishing a comprehensive evaluation model andanalyzing the evaluation model. By utilizing the relationship between color changes in the disease areas and wheat diseased panicle rate levels in the gradual pathologic process of wheat head scab, the color characteristic cloud of the disease areas is obtained by combining with three numerical characteristics of the cloud model, the wheat head scab levels are subjected to non-rigid diversion by utilizing the cloud model, and comprehensive evaluation is realized.

The invention discloses a menu rapid operating method based on a bi-rotation button, and the method is characterized in that a display interface is divided into a plurality of large groups according to the type and the function, all large groups form a main menu of a menu bar, each large group is provided with a plurality of submenus, each submenu corresponds to a relevant display interface, the interface is simple, visual and clear to arrange, the interface is reasonably classified, the corresponding interface display can be switched by rotating the bi-rotation button, and simplicity and convenience in operation can be realized; and the interface is intuitive and free from disorder, the reliability is high, and an interface needed by a pilot can be rapider and more accurate to find.

The invention relates to a method for evaluating a safety state of an elevated structure of urban rail transit. The method is used for performing overall safety state evaluation and grading on an elevated bridge of the urban rail transit. The method comprises the following steps of 1) determining an evaluation structure for evaluating the safety state of the elevated bridge of the urban rail transit, and corresponding evaluation indexes; 2) obtaining a technical condition score of a sub-item corresponding to each member of each part of the to-be-evaluated elevated bridge of the urban rail transit; 3) obtaining a safety state score of each corresponding part of the to-be-evaluated elevated bridge of the urban rail transit; 4) obtaining an overall safety state score of the to-be-evaluated elevated bridge of the urban rail transit; and 5) finally obtaining a safety state level of the to-be-evaluated elevated bridge of the urban rail transit, and designing maintenance measures according to the safety level. Compared with the prior art, the method has the advantages of comprehensive consideration, complete evaluation, reasonable grading, wide applicability and the like.

The invention discloses a food enterprise personalized risk grading management and control method and system, and relates to the field of food safety intelligent processing, and the method comprises the steps: building a food production and operation whole-process risk information base; establishing a risk grading evaluation model; collecting food enterprise data from a food production and operation whole-process risk information base; based on the food enterprise data and the risk grading evaluation model, obtaining a food enterprise risk grading result; and generating a corresponding management and control method based on the risk grading result of the food enterprise. The technical effects of reasonable and accurate risk grading of food enterprises and full and perfect element identification are achieved.

The invention relates to a method for determining an optimal value of the operation condition of a gas-steam combined cycle generating unit, and the method comprises the following steps: S1, building an online monitoring and economical operation system of the gas-steam combined cycle generating unit, and collecting the measuring point parameters of the unit; S2, carrying out the classification and screening of the operation conditions of the unit through employing a clustering algorithm, obtaining N stable operation condition classes, and distributing the measuring point parameters, which can reflect the operation condition of the unit, to different stable operation condition classes; S3, carrying out the data mining of the optimal values of the measuring point parameters for each stable operation condition class, obtaining the optimal value of each stable operation condition class, calculating the confidence of each optimal value, and determining the final preferred value of each stable operation condition class according to the confidence. Compared with the prior art, the method facilitates the improvement of economical operation level of the unit.

The invention relates to a garden waste treatment method and system. The garden waste treatment method comprises the following steps: S1, collecting: collecting waste in a garden; S2, sorting: dividing the collected waste into a high component category and a low component category according to content of organic substance components, and independently selecting out barks in the high-component category, wherein the high-component category comprises barks, tree leaves and grass cuttings, and the low-component category comprises branches and trunks; S3, classifying the low component category: scraping off the branches and the barks on the trunks, classifying the scraped-off barks into the high-component category, and conveying the branches and the trunks with barks scraped off to a wood processing plant; and S4, classifying the high component category: sequentially crushing, dyeing, fermenting and drying the barks in the waste of the high-component category to form improved type coverings, and composting the tree leaves and the grass cuttings in the waste of the high-component category. The garden waste treatment method has the effects of reasonably classifying the garden waste, increasing a resource utilization rate, and improving the treatment efficiency of the garden waste.

The invention relates to the technical field of sponge cities, in particular to a sponge city seepage drainage structure and a water circulation system. The sponge city seepage drainage structure andthe water circulation system comprise a hollow drainage well, the bottom of the drainage well communicates with a municipal drainage system, a first filter screen is arranged on the upper surface of the drainage well, a driving motor is arranged on the inner wall of the drainage well, a rotating shaft is coaxially arranged on an output shaft of the driving motor, a second filter screen is arrangedin the drainage well, the aperture of the second filter screen is smaller than that of the first filter screen, the rotating shaft is located between the first filter screen and the second filter screen, a hollow garbage well is arranged on one side of the drainage well, a discharge port is formed between the drainage well and the garbage well, one end of the second filter screen is located at the edge of the discharge port and lower than the other end of the second filter screen, a cover plate is arranged at the top of the garbage well, and a collecting box is arranged in the garbage well. The system can classify garbage in rainwater so as to indirectly improve the utilization rate of the rainwater, and therefore the effects of energy conservation and environmental protection are achieved.

The invention relates to an orthogonal walking mode self-carrying tray device and the purpose thereof; the device comprises a frame (2); the frame (2) is provided with a bearing plate (1); and the frame (2) is provided with a single motor driven walking mechanism and a single motor driven steering mechanism. The device has the advantages that the walking mechanism and the steering mechanism are respectively driven by a motor, so as to realize four-wheel synchronous walking and mechanism, and reduce unnecessary wear and power consumption; a power supply slipper supplies power through a safety slip line, outputs AC/DC and high-and-low-voltage power to supplies power to equipment; and the invention is applicable to matrix deployment circulating parking lots, electrical automobile charging stations, warehousing, logistics and the like, is flexible to schedule, convenient and quick, significantly reduces the dependency on artificial demand, particularly reduces the manpower resource cost in special warehousing environment, automatically complete the access operation of goods, automatically manages the goods in warehouses, so that the goods classification, transportation and warehousing are more reasonable.

The invention discloses a customer classification method and a customer classification system. The method comprises steps: information of all customers is acquired; a preset information field is selected from the information of each customer; a density-based clustering algorithm model is built, and according to the selected information field, the local density corresponding to each customer is calculated; and according to the calculated local density, all customers are classified to different classes. Accurate and comprehensive customer classification can be carried out, and an effective reference basis is provided for product promotion.

The invention discloses a clustering evaluation improvement method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a clustering sample of an advertisement business user, and constructing a clustering sample cluster according to the clustering sample; calculating distance values between the clustering samples and other clustering samples in the clustering sample cluster respectively, and forming distance vectors; normalizing the distance vector to obtain a weight vector and a weighted average value; obtaining a weighted average value of distances between the clustering samples and other clustering samples in the clustering sample cluster as a cohesion degree in the cluster; taking the weighted average value between the clustering samples and all clustering samples in the clustering sample cluster with the minimum distance as the separation degree of the clustering sample cluster with the minimum distance; and taking the difference between the separation degree and the cohesion degree as a molecule, taking the maximum value of the cohesion degree and the separation degree as a denominator, and taking a score value as an evaluation coefficient. The classification of advertisement service users is more reasonable, the user identification is improved, and the average weight effect of arithmetic mean values is improved.

The embodiment of the invention provides a test data construction method and a construction system, and the method comprises the steps: collecting test data from different sources, and enabling the test data to comprise interface data, flow data and case combination data; uploading the test data to a unified data resource pool for storage; and classifying the different test data according to a certain rule through machine learning, and combining the classified data obtained after classification to obtain new test data. By collecting and uniformly storing test data from different sources and classifying and combining different test data according to machine learning, new test data are obtained, multiple test data can be quickly and conveniently obtained, the coverage range is wide, and thepreparation efficiency and the coverage rate of the test data are integrally improved.

A traditional satellite resource management method is low in efficiency and cannot accurately monitor and display the state of satellite resources in real time. How to realize unified and efficient management of different satellite resources is always a problem puzzling the industry. The invention provides a method for acquiring and displaying a satellite resource state. Analyzing a logic hierarchical relationship of satellite resources; establishing a forwarding model and a communication resource model; according to satellite presentation and a frequency band presentation method, related information of satellite resources is displayed in a WEB page, satellite resource states are mutually associated and browsed through a WEB hyperlink technology, and the WEB page realizes data request andreal-time refreshing through an AJAX request and a Websocket technology. The forwarding model provided by the invention has good universality, and can be suitable for various processing forwarders such as transparent straight-through, transparent hinge and on-board exchange, on-board routing, frequency hopping and sub-band exchange.

The invention discloses a coking coal quality classification method. The method includes steps: 1) determining indexes having influences on coking coal cokeability, to be more specific, taking dry ash-free basis volatiles, determined Audibert-Arnu dilation degree b, vitrinite average maximum reflectivity, Kirschner's maximum mobility and coke optical texture structure as the indexes for determining coking coal cokeability, so that appearance limitation of existing classification indexes can be overcome, reasonability and accuracy in coking coal are achieved, application optimization is realized to give full play to functions of coking coal. In addition, by application of the Audibert-Arnu dilation degree index to the coking coal utilization technique and the compound coal utilization technique, coking coal quality characteristics are deeply mined, and optimal and reasonable resource application is realized.

The invention discloses a method for evaluating the integrated development level of urban and rural passenger transport, comprising the following steps: 1) selecting an integrated development model of urban and rural passenger transport; 2) establishing a structural composition for evaluating the integrated development level of urban and rural passenger transport, and determining the integrated development level of urban and rural passenger transport Evaluation structure and corresponding first-level evaluation indicators, the first-level evaluation indicators include facility construction evaluation, service effect evaluation and operation management evaluation; 3) respectively determine the second-level evaluation indicators under the first-level evaluation indicators and the second-level evaluation indicators The three-level evaluation index under the evaluation index constructs a complete evaluation structure; 4) Set the evaluation level of the evaluation index at each level and the evaluation standard level of the evaluation index corresponding to each level status; 5) Evaluate the evaluation indicators at all levels, and carry out Statistical analysis to determine the comprehensive evaluation value of the integrated development level of urban and rural passenger transport. The method for evaluating the development level of urban-rural passenger transport integration in the present invention is comprehensive in consideration, complete in evaluation, reasonable in classification and wide in practicability.

The invention discloses a mining method for a complex phosphate ore. The method includes a mining-process five-step method and a quality-control four-step method; the mining-process five-step method comprises the following steps of 1, grading, 2, separate mining, 3, separate transporting, 4 pile dividing, and 5, separate using; and the quality-control four-step method comprises the following stepsthat 1, geological grooving sampling is conducted in the grading process; 2, drilling sampling is conducted in the mining process; 3, template sampling is conducted in the separate transporting process; and 4, integrated sampling of a cargo yard is conducted in the separate sampling process. The mining method is suitable for the phosphate ore mountain with complex ore types, by means of the mining method, grading can be conducted on the phosphate ore, the phosphate ore of various types and grades is mined and can be directly provided to manufacturers of different needs, and the comprehensiveand efficient use of ore resources is achieved, so that the value of the ore is maximized.

The invention discloses a portable storage box for medical apparatus and instruments. The portable storage box comprises a storage box body, wherein a first storage box is arranged at the top of the storage box body; a first storage box cover is arranged above the first storage box; a second storage box cover is mounted on the left side of the first storage box cover; a second storage box is arranged below the first storage box; a third storage box is arranged below the second storage box; an ultraviolet light disinfection cabinet is mounted below the third storage box; a first opening and pulling groove is formed in the surface of the second storage box; a second opening and pulling groove is formed in the surface of the third storage box; locking devices are mounted on the right side of the second opening and pulling groove and the right side of the first opening and pulling groove respectively; third opening and pulling grooves are formed in the surface of the first storage box cover and the surface of the second storage box cover respectively; a smooth hinge is mounted on the left side of the first storage box; and idler wheels are welded at the bottom end of the storage box body. The portable storage box for the medical apparatus and instruments has the advantages of being reasonable in structure, convenient to carry, practical and high in efficiency.