35 results about "Learnability" patented technology

In predictive data mining, a process and tool presents a method to compare given competing algorithms to a derived reference, such as a baseline or benchmark. A result confidence as to the suitability of the competing algorithm to a given task is generated. In an exemplary embodiment, a randomized feature acting, simple, algorithm is used to generate the baseline. In an alternative embodiment, the process and tool is used to determine learnability of the given task. A mechanism to account for overfitting of data is described.

The invention discloses a natural language knowledge exploration system based on formal semantics reasoning and deep learning. The system comprises a machine learning module combining formal semantic reasoning and a machine learning method, and conducting machine learning containing semantics, an intention learning module learning a description intention of a to-be-explored natural language, a text subject extraction module analyzing a text subject content and paragraph content description intention via an LDA model, and a file structure classification model building module conducting model training based on a deep convolution neural network, automatically decorating and improving the neural network during the training, and building an automatic file structure classifier. Based on combination of automatic semantic reasoning, deep learning technology and natural language processing, non-structural data groups with documents and voices as representatives can be processed and reasoned in light of intentions, so real intention of verbal description can be learned; knowledge explored and retrieved can be used for new knowledge structure analysis and construction, so knowledge can be explored and retrieved in a way with high-efficiency, intelligence and learnability and self-evolution.

The invention discloses an intelligent maintenance auxiliary system for airborne equipment. The intelligent maintenance auxiliary system comprises an interaction module, an image detection and recognition module, a maintenance information management module, an analysis module and an augmented reality display module, The interaction module is used for obtaining input of a user, projection of a UI and interaction logic processing; The image detection and recognition module is used for detecting characters and specific patterns on the picture and recognizing content information; The analysis module is used for carrying out logic judgment and generating guided characters or 3D images; The maintenance information management module is used for sending the search request to the cloud database server; And the augmented reality display module is used for positioning a target position in a real environment in the field of view of the user and superposing the guidance information to the positionthrough an augmented reality technology. An augmented reality technology and intelligent maintenance assistance are organically combined, maintenance personnel are helped and guided to maintain airborne equipment in a natural interaction mode, and therefore operability and learnability of maintenance operation are improved.

The invention discloses a method and device for fatigue detection based on eye movement data. According to the method, through user interface (UI) interaction with a to-be-tested person, the response agility of the to-be-tested person is tested; while the test is performed, basic eye movement data is collected through an eye tracker to calculate eye movement fatigue characteristic data, and whether the to-be-tested person is in a fatigue state is judged finally through the response agility and the eye movement fatigue characteristic data in combination with a machine learning algorithm. The process of testing the response agility of the to-be-tested person through the UI interaction with a to-be-tested person is composed of a character search test and a reading comprehension test. Through testing, the detection accuracy rate of the method and device for fatigue detection based on the eye movement data reaches 95.2%; and compared with a conventional fatigue detection method, the method has the advantages of very high accuracy, practicality, simplicity, convenience and learnability.

The embodiment of the invention provides a method, a device, a receiving terminal, a transmitting terminal and equipment for group communication, wherein a group has a group label and is associated with the equipment, the group communication method comprises the steps: through second equipment, a second group of members receives communication data sent by a first group of members through first equipment, wherein the communication data includes data header and data content, and the data header includes a first group identification; and a contact person identification and the data content are displayed to the second group of members in the second equipment, wherein the contact person identification includes the first group identification. The embodiment can provide more diversified network behavior mode for a user, the learnability and operability of various network behaviors can be improved, and the flexibility and convenience of communication among groups can be enhanced.

The invention discloses a control system and method of a co-fusion robot for optical mirror surface machining. The intelligent control system controls a machining platform driving device, a cutting tool rack driving device, a flexible hydraulic supporting device and a rotating platform control device through a multi-axis motion controller and a machining platform motion controller respectively tomachine an optical mirror surface; a high-precision three-coordinate detector can feed information back to the intelligent control system in real time; a visual detection device collects spatial position information of a machined object and spatial position information of a mechanical arm in real time and feeds the information back to the intelligent control system. The intelligent control systemplans a machining route and conducts programming of a machining algorithm according to the feedback information in the machining process, the co-fusion robot has coordination performance and learnability at the same time in the process of machining the optical mirror surface, and therefore the requirement of optical mirror surface machining for high precision is met.

The invention relates to the technical field of data mining, and provides an oversampling method of an unbalanced data set. The method comprises the following steps: firstly, collecting an unbalanced data set, clustering the unbalanced data set based on a K-means method, and dividing the unbalanced data set into a minority class and a majority class according to the number of elements in the data set of each class; then, on the basis of an SMOTE method, carrying out oversampling on the minority class data set to obtain a synthesized minority class data set; then, performing oversampling with replacement on the synthesized minority class data set to obtain a new minority class data set, and forming a new data set; and finally, based on a CCA method, cleaning the new data set: clustering the new data set, calculating and sorting Euclidean distances between each sample in each class cluster and other samples in the class cluster, and deleting the sample corresponding to the farthest Euclidean distance to obtain the cleaned data set. According to the method, more minority class samples can be effectively synthesized, the learnability of the samples is improved, and the effectiveness of the samples is improved.

A 26 radical root Chinese and English harmonic inputting method is a pure code Chinese character inputting method which takes Chinese character common radical roots as the principal thing and takes 25 kinds of double stokes formed by combining every two of five kinds of single stroke as the auxiliary thing. Coding is conducted by adopting information such as the most specific structure, the common radicals and the strokes of Chinese characters, the single stroke, the 25 kinds of double strokes and the 26 common radical roots are regularly distributed on an English keyboard, each English word key corresponds to a unique common radical root, the effect that a user knows the roots by seeing the key positions and knows the key positions by seeing the roots is achieved, conditioned reflex is easy to form, and mechanical memorizing is not required. A coding rule is simple, and end pen identification codes do not exist. Few characters with identical codes exist in the first-level and second-level character base of the national standard, the dynamic coincident code rate is only about 0.5%, 98% of the Chinese characters can be input through a three-level brief code, and high-speed touch typing can be achieved slightly. The inputting method is easy to learn, difficult to forget and less in code coincidence and achieves perfect unification of inputting learnability and inputting high efficiency of the Chinese characters.

Disclosed is a thirty Chinese character component input method. The thirty Chinese character component input method is a pure shape-code input method with 30 Chinese common character components centered and 5 single strokes and 25 double stokes formed through pair combination of the 5 single strokes supplemented. According to the thirty Chinese character component input method, the information of Chinese characters such as the most concise structures, the common components and the strokes are encoded; for the 5 single strokes, the 25 double strokes and the 30 common components of Chinese characters, 30 encoding keys on an English keyword correspond to the common components one to one, and the layout of the keyboard is approximate to left-right symmetric distribution, so that the character component layout has strong regularity, and a user can know the keys once seeing the character components to achieve conditional reflex instead of mechanical memorizing. The encoding rules are simple without stoke end identification codes, few code-repeated characters in level 1 and level 2 international character libraries can be obtained, and the dynamic code repeating rate is as low as 0.3%; 98.5% of Chinese characters can be input through level-3 brevity codes, accordingly high-speed touch typing can be achieved easily. The thirty Chinese character component input method is easy to memorize and unforgettable, has few repeated codes and achieves complete integration of learnability and input efficiency of Chinese character input methods.

The invention discloses a push system of personal information and is related to the information technology field. The system includes a personal interest acquisition system, an information push system and an information acquisition system; a user end includes a data receiving module, a user registration module, a user searching module and a user modification module; the personal interest acquisition system includes a user attribute static analysis module, a word frequency statistics module, a time analysis module, a log mining analysis module, an analysis module and a personal interest set; the information push system includes an information retrieval module, an interest acquisition module and an information sending module; the information acquisition system includes an information acquisition module, an information classification module and a data base. The system provides an information push system according to the personal interest and is of high accuracy, high learnability and strong pertinence.

The invention discloses a tri-circle concentric laser locating device which comprises a high-precision laser pointer. The high-precision laser pointer is fixed in a fixing barrel center hole of a laser pointer fixing barrel. The periphery of the laser pointer fixing barrel is provided with a protective cover of the laser pointer fixing barrel, the tail of the protective cover of the laser pointer fixing barrel is in concentrical fixed connection with a first main circular plate, the tail of the laser pointer fixing barrel is provided with a fixing barrel threaded hole, and the fixing barrel threaded hole is in threaded concentric connection with the first main circular plate through a threaded fastener. The circle center of the first main circular plate is provided with a center hole of the first main circular plate, and the circle center of the first main circular plate is coincided with the circle center of the fixing barrel center hole. The head of the laser pointer fixing barrel is provided with a second main circular plate, the circle center of the second main circular plate is provided with a center hole of the second main circular plate, and the circle center of the center hole of the second main circular plate is coincided with the circle center of the fixing barrel center hole. The device has the advantages of good stability, high safety and reliability, operation simplicity, learnability, manuverability, compact and simple structure, moderate weight, high precision, low manufacture cost and low maintenance fee.

The name of the invention is a simplified Chinese characters spelling stroke joint input method, which is abbreviated as the joint input method and belongs to the field of computer information processing. The invention is a set of brand-new keyboard input method specific to simplified Chinese characters used by a computer, aiming at providing the input of simplified Chinese characters with the characteristics of convenience and learnability as well as fast input, thus overcoming the drawbacks of the existing input method which is either easy to learn but slow in speed or can achieve rapid speed but is hard to learn and use. Simply speaking, the joint input method utilizes the initial and characters spelling initial in a plurality of characters as well as corresponding key names with a plurality of strokes to form the coding with full code and four digits, and 26 common letter keys of a keyboard are correspondingly used to input Chinese characters, namely spelling and strokes are combined to be used. In addition, a brevity code can be arranged sequentially to further improve the input speed and reduce the repeated codes simultaneously.

The invention discloses a first-letter-spelling Chinese character input method. The first-letter-spelling Chinese character input method comprises the following steps: directly linking a first later of a Chinese character or a first letter of a radial first letter with a universal keyboard; linking the radical which is not the Chinese character with the keyboard according to rules described in a dictionary. The first-letter-spelling Chinese character input method has the advantages that the accuracy is high as a repeat code rate of the first letter of spelling is 38.9% and a page turning rate is 2.3% by coding 4000 Chinese characters one by one; learnability is realized as vowels and tones are not needed to be used, the input method is easy to learn; internationalism is realized as only the first-letter pronunciation of the Chinese character is needed to be known for the first letter of spelling, and the input method is almost the same for a simplified Chinese character and an original complex form of a simplified Chinese character which are in the same in pronunciation; public benefit is realized as the first letter of spelling is used for typing, and therefore knowledge of a structure, an order of strokes and the like of the Chinese character can be strengthened; expansibility is realized as in addition to applying to a computer, the first letter of spelling also can be expanded, applied and pre-installed to mobile equipment (a mobile phone, a tablet computer, and the like).

The invention relates to a tetra stroke code Chinese character input method, which integrates radical, stroke and stroke shape information of tetra number of Chinese character into a whole. The tetra number at the upper left corner of the radical and an end stroke are dealt with code-bit, coded according to 201 radicals and about 300 etymons of the newest edition 'the Uniform Radical table of Chinese Character', the radical is the indexing radical of a tenth version Xinhua Dictionary. The tetra stroke code contains four input methods of four-character, trigram code, first and last code, and digital code, and the tetra number and the stroke code can be used for inputting characters at the same time. The tetra stroke code has the advantages of easy coding, compatible character code and digital code, uniform complex font and familiar style, standard, learnability, usability, being not necessary to memorize the etymon or identify voice, identifying code at sight of characters, etc., and lowers the difficulty from learning, memory and operation of fractal code to the minimum. Only on the condition of the function of consulting a dictionary, Chinese characters can be input into a computer or mobile phone quickly by the method. The invention is suitable for junior and primary students, officials, and the mid-aged and old-aged, and is also suitable for the junior and primary students to master the basic indexing method of a Chinese character dictionary, for example standard radical, stroke and the tetra number, etc.

The invention discloses an indoor child shooting tent. The tent comprises a tent cloth and support rods; the tent cloth comprises 3-6 side cloths which are mutually connected through sewing, one of the side cloth is provided with a tent cloth, at least one side cloth is provided with a shooting opening, one of the side cloth is provided with a shooting device, the shooting device comprises two symmetrically arranged mounting cloths, a shooting frame, a connecting frame and hook-and-loop fasteners, multiple adjusting strips arranged at intervals are connected to the mounting cloth through sewing, a mounting rod is installed in a mounting pocket of each mounting cloth, the mounding rods are located between the mounting cloths and the adjusting strips, the shooting frame is connected with theconnecting frame, and the two ends of the connecting frame are connected to the mounting rods through connecting ropes; the bottom face of the connecting frame is connected with the tent cloth through the hook-and-loop fasteners. According to the indoor child shooting tent, installation of a shooting basket and the tent is stable, the basket is not likely to incline, children shooting is accurate, multiple persons can play at the same time conveniently, the shooting modes are diverse, and the entertainment and learnability of the tent are enhanced.