58 results about "Human assessment" patented technology

Systems and methods are provided for analyzing a history of meetings, the attendees, date of occurrence, and other content to determine the value of the meetings and the attendees. The importance of people and the meetings they attend can be derived based on patterns of attendees. In one embodiment, the meta-data of meetings and the attendees can be used to determine value without requiring time-consuming manual steps or manual evaluation of people and their titles. A graph of meetings and its attendees can be generated and used by one or more automated software agents to place value to the content of the meeting, its agenda, and other meeting collateral such as meeting briefs / attachments of meetings. Accordingly, embodiments dramatically reduce the need for human examination of meeting history.

The invention provides a Knowledge graph-based resume evaluation method, apparatus and device. According to the embodiment of the invention, the resume knowledge graph is generated by utilizing the evaluated historical resumes in advance, and the resume knowledge graph comprises the correlation information of the resume information of the historical resumes relative to the post requirements; aiming at the unstructured information in the to-be-evaluated resume, the semantic components of the unstructured information in the to-be-evaluated resume can be labeled by utilizing the resume knowledgegraph; a first structure label capable of representing the unstructured information is obtained; correlation information of the unstructured information relative to the post requirements is obtained from the resume knowledge graph through the first structure label, the correlation information serves as the evaluation factor to determine the correlation of the to-be-evaluated resume relative to thetarget post, unstructured information in the resume can be understood and analyzed and serves as an important basis for evaluating the resume, the evaluation quality is improved, and human resource waste caused by manual resume evaluation is avoided.

The present invention relates to a video analysis-based newborn pain expression identification method. According to the present invention, by introducing a deep learning method based on a three dimensional convolutional neural network and applying to the newborn pain expression identification work, the expressions when the newborns are in the peaceful and crying states and the expressions, such as the mild pain, the severe pain, etc., caused by the painful operations can be identified effectively, and a new path is provided for developing a newborn pain automatic assessment system. Concretely, by introducing the three dimensional convolutional neural network, and using a 3D convolution kernel to extract the time domain and spatial domain features of the video fragments, the tedious operation of extracting the features artificially is avoided, the extracted features are preferable, and the newborn pain expressions are learned and identified automatically on a deep learning platform. Compared with a conventional artificial assessment mode, the video analysis-based newborn pain expression identification method is more objective and accurate, and saves a lot of manpower resources.

A method for ultrasound diagnosis includes determining a first risk of malignancy based on a human assessment of a first set of features of one or more ultrasound images of a target; determining a second risk of malignancy based on an automatically extracted second set of features of the one or more ultrasound images; determining at least one overall risk value based on the first risk of malignancy and the second risk of malignancy; and characterizing the at least one overall risk value as one of a high confidence assessment or a low confidence assessment.

The invention discloses a quantifiable enterprise development situation evaluation model. Two first-level evaluation indexes, which are the innovation capacity and introduction of talents, are added into an index system of the model. Evaluation of the model for enterprise development situations reflects important inclination to the evaluation direction of enterprise soft power and future development potential, and the scientificity and the dynamic nature of evaluation tools, and persuasion and policy directivity of evaluation results are improved; various indexes are stratified and uniformly quantized, and a final evaluation result is attributed to a determinate scalar quantity, thereby being capable of providing a scientific and objective basis for decision making for decision makers of relevant departments; and compared with previous manual evaluation, the evaluation model can evaluate the development situations of enterprises accurately and objectively and reduces interferences of subjective factors to the maximum degree.

The embodiment of the invention discloses a method and device for evaluating relevance of retrieval texts, a server and a storage medium. The method comprises the following steps: performing text features extraction on a plurality of sample pairs composed of query and retrieval texts, wherein the text features include original text features and structured text features; and taking the text features and correlation labels of the plurality of sample pairs as corpuses and training, and obtaining an evaluation model, wherein the evaluation mode is used for evaluating the relevance between the query and retrieval texts. According to the embodiment of the invention, when evaluating the relevance of the retrieval texts, the depth and automation of evaluation problems are taken into account, and the accuracy and generalization recognition capability of the decision logic are taken into account; besides, the comprehensiveness and the matching degree of retrieval recall can be improved through evaluation, and user experience can be improved; and meanwhile, the training and use of the evaluation model make the coverage of evaluated texts large and the cost of manual evaluation is reduced.

The invention is applicable to the technical field of big data processing. A method for evaluating security risk based on big data and terminal device are provided. The method comprises: receiving a security request sent by a client, the security request containing the customer's identity and policy change information, obtaining the customer history transaction information corresponding to the customer identity, and determining the customer 's reputation index according to the historical transaction information, identifying the policy change items in the policy change information, obtaining the project category of the policy change items, obtaining the risk value of each project category, importing the risk value into the preset risk model to obtain the risk index, and analyzing and evaluating the preservation request based on the reputation index and the risk index. The manual audit is not required, the embodiment of the invention can save the human cost of evaluating the preservationrisk, improve the evaluation efficiency, eliminate the influence of the subjective factors of the manual evaluation, and improve the accuracy of the evaluation.

A remote deposit capture system is provided that includes the means for human evaluation of financial transactions, and well as the application of rules based methodologies.

The invention provides an insurance client health condition analysis method. The method comprises the steps of obtaining a plurality of preset type data of insurance clients, and generating a health risk level analysis model based on the preset type data of insurance clients and according to a preset model generation rule; upon receiving a health risk level analysis instruction for a to-be-insured client, receiving the insurance data of the client; acquiring preset type data from the insurance data of the client, adopting the preset type data as health risk data and determining risk factor data corresponding to each preset type data; substituting the risk factor data corresponding to each preset type data into the health risk level analysis model and analyzing a health risk level corresponding to the client. The invention also provides a server suitable for the above method. Based on the invention, the health risk level of an insurance client can be quickly evaluated. Therefore, a great amount of manpower and material resources, consumed during the traditional manual evaluation process, can be avoided.

The invention is applicable to the field of computers, and provides a credit risk evaluation method and system, computer equipment and a storage medium. The credit risk evaluation method comprises thefollowing steps: collecting communication relationship information of a target user; constructing a communication relationship graph database of a target user, wherein the communication relationshipgraph database takes a communication account in a communication relationship as a node; and risk characteristics of associated nodes in the communication relationship graph database are mined, a riskevaluation result of a target user is determined according to the risk characteristics of the associated nodes, and the associated nodes are nodes directly or indirectly associated with target user nodes in the communication relationship graph database. According to the scheme, the risk evaluation result of the target user can be determined according to the risk characteristics of the associated nodes by establishing the communication relationship graph database of the target user and mining the risk characteristics of the associated nodes in the communication relationship graph database, manual evaluation is replaced, and the evaluation efficiency and the reliability of the evaluation result are improved.

Systems and methods for automatic evaluation of the quality of NLG outputs. In some aspects of the technology, a learned evaluation model may be pretrained first using NLG model pretraining tasks, and then with further pretraining tasks using automatically generated synthetic sentence pairs. In some cases, following pretraining, the evaluation model may be further fine-tuned using a set of human-graded sentence pairs, so that it learns to approximate the grades allocated by the human evaluators.

Particular aspects provide methods for enhancing customer satisfaction and / or loyalty with respect to customer-driven entities, and more particularly to novel methods for improving customer satisfaction and / or loyalty by enhancing the emotional competence of entity personnel, and by enhancing learning transference or application of the enhanced emotional competence to customer-contact job-tasks. Particular aspects comprise: identifying a customer-driven entity having participant personnel having customer-contact job-tasks; assessing the customer-contact participants, using an emotional competency assessment tool and at least one Human assessment tool expert; participating, by the customer-contact participants, in contract learning activity suitable to provide for learning of enhanced emotional competency skill or proficiency; and applying, by the customer-contact participants and with facilitation by a Human expert, the enhanced skill or proficiency to customer-contact job-tasks, wherein establishing enhanced customer satisfaction and customer loyalty is afforded. Additional aspect provide assessment kits for practicing the inventive methods.

The invention discloses a second-hand house evaluation system based on machine learning. The system comprises at least one group of user side, an input module, a collection module, a database, a calculation and evaluation module and an artificial evaluation module, and the input module inputs real estate information and sends the real estate information to the collection module. The collection module captures keywords of the real estate information, the real estate information is classified and sent to the database, the database is stored in the cloud server, the calculation and evaluation module compares the input real estate information for mechanical evaluation, and the artificial evaluation module adjusts the mechanical evaluation and verifies the authenticity of the input real estateinformation at the same time. Mechanical evaluation of the second-hand house is realized by combining data capture classification and a mechanical algorithm, and artificial evaluation is added to ensure accurate and reasonable price and real and reliable real estate.

The present invention provides a human-computer interactive rehabilitation system, which can automatically calculate rehabilitation strength suitable for the patient, so that it is not necessary to manually evaluate and adjust the parameter settings in human-computer interactive rehabilitation system when different patients use it. At the same time, the human-machine interactive rehabilitation system and the hospital end can track the rehabilitation status and intervene through the data platform at any time. The platform establishes a cloud community feedback and encouragement mechanism, and immediately transmits the rehabilitation results to the designated barriers of the patients, provides patient encouragement feedback, and strengthens the community interaction and linkage in the medical relationship.

The invention discloses a network communication platform-based Yellow River dam bank monitoring and alarm system. The system comprises a monitoring terminal, a monitoring center, a case comparison analysis database, an online evaluation module and an alarm signal receiving terminal. According to the system, a plurality of modules such as monitoring, danger check and danger report modules are organically combined together; a highly-visualized interface is provided, so that data can be read and understood easily; the dangerous conditions of the Yellow River can be mastered comprehensively; and the case analysis database and the online automatic or manual evaluation module are adopted, emergency handling sites and engineering conditions can be analyzed timely, and therefore, strong information support can be provided for scientific, timely and accurate decision-making in the leadership of flood control and emergency handling.

The invention is suitable for the technical field of computers, and provides a risk control method and device for a vehicle insurance claim settlement, and the method comprises the steps: obtaining the related data information of a to-be-predicted vehicle and the external factor data corresponding to the related data information; processing the related data information of the to-be-predicted vehicle and the external factor data of the to-be-predicted vehicle by adopting a residual attention network, an auto-encoder and a neural network to obtain a vehicle loss evaluation index of the to-be-predicted vehicle; determining a subsequent vehicle insurance claim settlement process according to the vehicle loss evaluation index of the to-be-predicted vehicle, and simulating the use of a vehicle loss evaluation method based on artificial intelligence, thereby guaranteeing the accuracy of an evaluation result; the unreasonable risk of a subsequent claim settlement program is remarkably reducedaccording to an accurate evaluation result; adverse consequences or influences caused by insurance fraud or manual evaluation are reduced; and the benefits of insurance companies and owners of vehicles involved in cases are ensured.

The invention discloses an education and teaching evaluation system based on an artificial intelligence blackboard and an evaluation method thereof, and relates to the technical field of big data analysis. The system comprises a database, a data acquisition unit, a query statistics unit, a comprehensive evaluation unit and a system maintenance unit; the data acquisition unit comprises a data acquisition module, a data preprocessing module and a multi-source conversion module; the query statistics unit comprises a query module and a statistics module; the comprehensive evaluation unit comprisesa decision tree module and a classification module; and the system maintenance unit comprises a user management module and a data management module. A large amount of data related to education and teaching and corresponding evaluation results are recorded through the system database, the decision tree algorithm is adopted to associate behaviors and results to construct the classification model, and the evaluation results are made for the preprocessed classification data according to the classification model, so that time and labor consumption of manual evaluation is avoided, and the educationand teaching evaluation accuracy and efficiency are improved.

The invention discloses a hardware health evaluation method of a database server based on an expert model. The evaluation method includes collecting operating system hardware indexes, dividing the hardware indexes into indexes in expert model, assigning score weights to the indexes in expert model, and establishing multi-level thresholds and deduction standards, wherein the index values fall within different threshold ranges and deduct corresponding grade scores; automatically and periodically collecting hardware indicators; according to the rules of expert model, the hardware index of database server is calculated, and the health score of database server hardware is output. The above evaluation method selects the key indexes of database hardware and solves the reliability problem of indexselection. Combining and matching the indexes, a database hardware health model is constructed, and the problem of singularity of evaluation is solved by assigning weights to the indexes in the modeland setting dynamic thresholds. Through automatic collection and model calculation, the hardware health degree of database is evaluated, and the problem of high cost of manual evaluation is solved.

A system allows a provider to better monitor the health of customer accounts. The present system monitors customer utilization and adoption of their product using machine data along with human evaluation data. The customer may be monitored in several areas of usage with a product. The monitoring results include machine data (usage data) which is combined with human input to generate a health score for a customer. Once the health score is determined, action items may be assigned, a renewal possibility may be considered for future business, and expansion possibilities may be determined.

The invention discloses a method of interactive reinforcement learning from demonstration and human assessment feedback. An IRL-TAMER is formed by combining an inverse reinforcement learning IRL and aTAMER framework. The method has the beneficial effect that the intelligent agent can effectively learn from human rewards and demonstration through the learning method.

The invention discloses a depression testing and treating method. The method is characterized in that firstly, a psychological CT (computerized tomography) system is used for testing, the problems of differences and high misdiagnosis rate of test results through human assessment are solved, and the objectivity and accuracy of test results are guaranteed. Parameters of a transcranial magnetic therapy instrument and a transcranial direct current electrical stimulator are automatically generated according to a detection report of the psychological CT system, the basis and objectivity of the generated data are high, and the treatment accuracy and safety are guaranteed. A biofeedback therapy instrument is adopted finally for physiological adjustment of a patient with the depression, physical and psychological treatment is realized, and self-knowledge is increased.

The invention discloses a power transmission line aging degree evaluating method which comprises the steps of acquiring production system data, historical testing data and environment meteorological data; according to the physical and chemical testing data, electrical testing data and mathematical research accomplishment of a laboratory transmission line sample, establishing a power transmission line aging degree evaluation model; acquiring real-time production system data and environment meteorological data of the power transmission line; obtaining a preliminary condition of the power transmission line aging degree through the evaluation model; performing physical and chemical testing and electrical testing on the power transmission line and acquiring latest testing data of the power transmission line; and according to the evaluation model, performing further calculation for obtaining the specific condition of the power transmission line aging degree. Compared with a traditional artificial evaluation method, the method according to the invention can objectively and accurately reflect the aging degree of the power transmission line, and a relatively fuzzy aging degree which is obtained through observation and experience in traditional power transmission line aging degree evaluation is prevented, thereby realizing objective and accurate aging degree evaluation to the power transmission line.

In the embodiment of the invention, asset information of a target object is determined, and target assets in the asset information of the target object are marked, wherein the target assets comprise one or two of entity assets and intangible assets; asset evaluation reference data are acquired, asset evaluation rules are determined according to the asset evaluation reference data, and the asset evaluation rules are used for determining asset evaluation parameters corresponding to different asset categories; and according to the asset evaluation rule, the marked target asset is evaluated to obtain a current asset value of the target asset. Thus, the target assets can be automatically marked from the asset information of the target objects, the assets are automatically evaluated based on the asset evaluation rules, and compared with an existing asset management mode adopting manual evaluation, the asset management automation degree can be greatly improved, the asset management efficiency is improved, errors are reduced, and consumption of manpower and material resources is reduced.

The invention provides a building post-earthquake damage grade evaluation method based on deep learning, relates to the field of earthquake engineering, and is characterized in that damaged and damaged house images in an earthquake area are acquired, marked and a data set is established, and the buildings in the images are subjected to earthquake damage classification based on deep learning. According to the method, accurate damage classification can be quickly and efficiently performed on the building in the earthquake-bearing area, the calculation capability and the quick decision-making capability of the deep neural network can be utilized to replace on-site manual evaluation, time and energy consumption is greatly reduced, and under the condition that the accuracy is ensured, the damage condition of the building in a large-range earthquake-bearing area is efficiently judged.

The invention relates to an abandoned factory availability evaluation method and system based on computer vision, and belongs to the technical field of computer vision. The method comprises the following steps: obtaining an appearance integrity grade of a waste factory, the number of internal equipment of the waste factory, an equipment integrity grade of the waste factory, a green land area and a water area in a neighborhood range of the waste factory, the number of public transport stations in the neighborhood range of the waste factory, the number and the length of roads in the neighborhood range of the waste factory, the number of living facilities of preset types in the neighborhood range of the waste factory, the area of buildings in the neighborhood range of the waste factory, and the distance from the center of the waste factory to the center of a city, so as to obtain all of the indices; according to the indexes and the importance degree values of the indexes in the preset transformation types, acquiring the evaluation values corresponding to the different preset transformation types. According to the method, the availability of the abandoned factory can be objectively and accurately evaluated, and the subjectivity and low efficiency of manual evaluation are avoided.

The invention discloses a method and a device for monitoring website visitor values. The method comprises the following steps: monitoring visitor information of a website, and according to the pre-divided grades of the website visitor values, acquiring the probability of occurrence P(cj) of each grade cj in a data sample, wherein j=1,2,...,N, and N is the number of the divided grades; after acquiring the visitor information X, further acquiring the dependent probabilities P(X) and P(X|cj) of the visitor information in the data sample; and calculating P(cj|X)=(P(X|cj)P(cj) / P(X)), (j=1,2,...,N), evaluating the visitor values by using the calculation result, and monitoring and outputting the evaluating result of the visitor value in real time. Because the processing of the scheme takes historical data samples as a basis and the historical data samples can effectively reflect the rule of an artificial evaluation, by applying the scheme to evaluate the visitor value, an evaluation result almost accordant with that of the artificial evaluation can be obtained.

The invention provides an image quality evaluation method and device, a terminal and a readable storage medium, and relates to the technical field of computers. The method comprises the following steps: determining target dimension data of a target image, wherein the target dimension data comprises one or more of image definition dimension data, color richness dimension data and value degree dimension data; and according to the target dimension data, determining a quality score of the target image, the quality score being used for evaluating the quality of the target image. Target dimension data are comprehensively considered under one or more dimensions; the quality score of the target image is determined to evaluate the quality of the target image, the evaluation of the target image is more objective, the problems of poor accuracy and low speed of manual evaluation are solved, the accuracy and speed of image quality evaluation can be improved, and the method can be suitable for evaluating a large number of images.

Embodiments of the invention provide an enterprise credit assessment method and an enterprise credit assessment device. The method comprises the following steps: a server gets the information corresponding to an enterprise; the server queries missing main reference information in the enterprise registration information, and subtracting a score corresponding to the missing main reference information from the initial value of a score corresponding to the enterprise; the server queries whether the enterprise registration information contains auxiliary reference information in addition to the main reference information, and adds a score corresponding to the auxiliary reference information to the score corresponding to the enterprise; and the server determines the credit level of the enterprise according to the result of comparison between the score corresponding to the enterprise and a preset threshold, wherein, the credit of the enterprise is at a first level when the score corresponding to the enterprise is greater than or equal to the preset threshold, the credit of the enterprise is at a second level when the score corresponding to the enterprise is smaller than the preset threshold, and the first level is higher than the second level. The method avoids the subjectivity of human assessment, makes cheating impossible, and makes assessment more objective and reliable.

The invention relates to a relay protection differential intelligent inspection and anomaly discrimination method and a technical system. The method comprising the steps of uniqueness tagging and labeling of a relay protection device, differentiated inspection template configuration, inspection range delineation based on an equipment owner, automatic identification of the relay protection device through code scanning, analysis and extraction of a type manufacturer of the relay protection device, differentiated intelligent inspection, automatic judgment of abnormal data and automatic generation of an inspection report. The differentiated intelligent inspection method is provided for complex inspection management of relay protection, efficient intelligent inspection and data acquisition of devices of various protection types are achieved, the inspection efficiency and accuracy are greatly improved, and the safe and stable operation level of relay protection is effectively improved; meanwhile, the workload and the error rate of manual evaluation, analysis and judgment are reduced, and the accident occurrence probability is greatly reduced.

The invention provides a depression tendency assessment method and device. The method comprises the following steps: obtaining response time of a tested person making an emotional decision; performing feature extraction on the reaction time to obtain original feature parameters; performing dimension reduction processing on the original feature parameters to obtain evaluation feature parameters; and inputting the evaluation feature parameters into a pre-trained extreme learning machine model to obtain a depression tendency index. According to the method, depression tendency evaluation is performed according to the reaction time of the emotional decision made by the tested person, so that the attention on the emotion and the reaction speed can be increased. Compared with manual evaluation in the related technology, influence of subjective factors is reduced, and compared with evaluation through physiological signals in the related technology, environment and equipment noise is reduced, and improvement of depression tendency evaluation accuracy is facilitated.