CN113076459A - Neural network building method and system based on AI consultation - Google Patents
Neural network building method and system based on AI consultation Download PDFInfo
- Publication number
- CN113076459A CN113076459A CN202110459925.2A CN202110459925A CN113076459A CN 113076459 A CN113076459 A CN 113076459A CN 202110459925 A CN202110459925 A CN 202110459925A CN 113076459 A CN113076459 A CN 113076459A
- Authority
- CN
- China
- Prior art keywords
- neural network
- module
- virtual
- database
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 90
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000009193 crawling Effects 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 15
- 230000005540 biological transmission Effects 0.000 claims abstract description 13
- 238000004364 calculation method Methods 0.000 claims abstract description 13
- 230000001953 sensory effect Effects 0.000 claims abstract description 11
- 239000003999 initiator Substances 0.000 claims abstract description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 36
- 238000004519 manufacturing process Methods 0.000 claims description 9
- 238000012552 review Methods 0.000 claims description 9
- 230000000007 visual effect Effects 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 7
- 239000002689 soil Substances 0.000 claims description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 7
- 210000001503 joint Anatomy 0.000 claims description 5
- 230000008447 perception Effects 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 238000003860 storage Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 230000004931 aggregating effect Effects 0.000 claims description 2
- 238000010276 construction Methods 0.000 claims 1
- 238000004458 analytical method Methods 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 230000008054 signal transmission Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 210000003792 cranial nerve Anatomy 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000000638 stimulation Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Neurology (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a neural network building method and system based on AI consultation, which comprises a neural network system and a virtual system, wherein the neural network system is connected with the virtual system through edge calculation, the neural network building method comprises the steps of preparing a database, extracting characteristics as input to the neural network, and calling and storing information content sent by an information initiator and collected content into an information content database by a database interface module when the neural network is built. The invention ensures that signals can be stably transmitted when data update information is carried out, improves the transmission efficiency, can avoid entering an endless deep branch, and can enable a web crawler to carry out parallel processing, thereby crawling required content, accelerating crawling speed and efficiency, removing some illegal information, then displaying the structure, avoiding that all crawling results cannot be displayed after illegal content appears, effectively ensuring the diversity of sensory signals and ensuring the accuracy of the signals in the system by collecting the signals in multiple aspects.
Description
Technical Field
The invention relates to a cloud computing building method, in particular to a neural network building method and system based on AI consultation.
Background
The intelligent system is realized according to a deterministic algorithm, the processing result is a deterministic system, the artificial intelligence is that a computer is used for realizing the human mind function, namely, the effect generated by the human mind is realized through the computer, the problem to be solved is that the neural network algorithm mainly carries the information of the target function on the parameter of massive weighted value through training, the weighted value W and the threshold value T need to obtain the best solution in the learning process, all the states need to be tested, the total times to be combined is { (W multiplied by T) n } multiplied by P, wherein n is the number of nodes of the neural network in one layer, P is the number of layers of the neural network, the high-index calculation complexity causes huge calculation amount, the needed hardware overhead is huge, and the probability gradient descent method adopted by the loss function of the deep learning effect is called SGD for short, the obtained training value is a local optimal solution, so that the problem of 'black box' inevitably occurs, the threshold value in the model of the neural network belongs to artificial definition, the mechanism of the neural network of the brain of a human is greatly different, the principle of the stimulation signal of the cranial nerve cannot be fully embodied in the traditional neural network model, the mechanism that the head of the human judges differently according to the different excitation degrees generated by the neural signal of a neuron cannot be embodied in the current neural network model, and the like, the current neural network model can only be academic, represents a directional theory, and has a large level difference with the practical application. At present, in the stage of deep learning, compared with the traditional neural network, the number of hidden layers is increased, which further increases the complexity of calculation, although some optimization algorithms are introduced in the learning, the optimization algorithms do not depart from the basis of the original neural network, the fatal problem of the traditional neural network cannot be solved, and the prospect of wide application is difficult to expect.
Disclosure of Invention
The invention aims to provide a neural network building method and system based on AI consultation, which have the advantages of ensuring that signals can be stably transmitted when data update information is carried out, improving transmission efficiency, avoiding trapping in an endless deep branch, enabling a web crawler to carry out parallel processing, crawling required contents, accelerating crawling speed and efficiency, displaying structures after some illegal information is removed, avoiding that all crawling results cannot be displayed after illegal contents are generated, effectively ensuring the diversity of sensory signals and ensuring the accuracy of signals in the system through multi-aspect signal acquisition, and solving the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme: an AI consultation based neural network building method comprises the following steps:
s1: preparing a database, extracting characteristics, inputting the characteristics into a neural network, calling and storing information content and collected content sent by an information initiator into an information content database by a database interface module when the neural network is built;
s2: building a neural network system, connecting a crawler algorithm in the system with the Internet of things and the Internet through a database interface module after the neural network system is built, then performing crawler for multiple times, calculating through the crawler algorithm, and transmitting the result to a database for storage through the database interface module;
s3: the neural network system is arranged in the Internet core server, the neural network system is in butt joint with the virtual system through the computing module, and the Internet core server interacts with a demander through a human-computer interaction interface on the display module;
s4: the crawling answer quoting module in the neural network system quotes and feeds back the information and content crawled by the crawler algorithm, the information and content are displayed by the display module, and the verification and retrieval are carried out by the manual review module.
Furthermore, the database in S1 is used for correcting and storing a plurality of groups of the crawled answers, the database interface module is used for connecting the neural network system with the internet of things and the internet, and the neural network system is internally provided with a detection system for detecting the interface problems of the database interface module.
Furthermore, a processing module for sensing whether the plurality of database interface modules are accessed with signals is arranged in the detection system, a signal sensing abnormity reminding module is arranged in the processing module, and when a certain abnormal database interface module is accessed with signals, a reminding interface is popped up through the signal sensing abnormity reminding module to ensure stable transmission of the signals; the signal sensing abnormity reminding module can be used for a plurality of database interface modules to sense and discharge faults.
Further, the web crawler of the crawling algorithm in the neural network system in S2 starts with a selected hyperlink, and proceeds with accessing one link according to one line until reaching a leaf node of the line, that is, an HTML file not containing any hyperlink, and after processing the line, proceeds to the next start page, and continues to access one of the links contained in the new start page until reaching the leaf node.
Furthermore, webpage data crawled by crawler nodes of the crawler algorithm can be stored in a resource library, and the resource library analyzes the crawled data and establishes indexes.
The invention also provides a technical scheme that: a neural network building system based on AI consultation comprises a neural network system and a virtual system, wherein the neural network system is connected with the virtual system through edge calculation;
the system comprises a neural network system, a simulation module and a display module, wherein the neural network system comprises a database interface module, a crawler algorithm, a calculation module, a crawl answer reference module, an artificial rechecking module and a display module, the database interface module is connected with the neural network system, the crawler algorithm and the crawl answer reference module are in butt joint through a database, the calculation module is used for butt joint with a virtual system, the artificial rechecking module rechecks the crawl answer reference module, and the display module can display the neural network system on;
the database interface modules are provided with a plurality of groups for the gap connection of different output ports, so that the stability of information transmission is ensured;
the manual rechecking module is used for labeling and classifying the data, rechecking the contents by adopting a plurality of paths and rechecking the results of the contents by aggregating the paths;
and the crawl answer quoting module is used for combing and quoting evidence-making for the information crawled by the crawler algorithm.
Furthermore, the virtual system comprises a virtual auditory system, a virtual visual system, a virtual sensory system and a virtual motor system, wherein the virtual auditory system, the virtual visual system, the virtual sensory system and the virtual motor system are all connected with the neural network system;
the virtual auditory system is composed of a plurality of audio acquisition modules, and stores and translates audio contents into characters for facilitating crawling in later period by collecting audio information acquired by the audio acquisition modules;
the virtual vision system is composed of a plurality of video acquisition modules, acquires video information, disassembles the video information into audio content and frame capturing content, and then screens and checks the audio content and the frame capturing content through an audio checking module and a picture checking module respectively;
the virtual sensing system comprises an air sensor, a water system sensor and a soil sensor, wherein the air sensor, the water system sensor and the soil sensor are all connected with a collecting module in the virtual sensing system, and collected sensing signals are transmitted and stored in the collecting module.
Furthermore, the virtual motion system comprises office equipment, household equipment, a cloud robot and production equipment, and the office equipment, the household equipment, the cloud robot and the production equipment are all connected with the internal database connected with the neural network system.
Further, the terminal for storing and installing the neural network system and the display module includes a mobile terminal such as a desktop, a collection, a notebook, etc.
Compared with the prior art, the invention has the beneficial effects that:
the neural network building method and system based on AI consultation ensure stable signal transmission by popping up a reminding interface through a signal perception abnormity reminding module, wherein the signal perception abnormity reminding module can be used for sensing a plurality of database interface modules and discharging a failed database interface module, so that the signals can be stably transmitted when data is updated, the transmission efficiency is improved, a web crawler can first grab all web pages including links in an initial web page, then select one of the linked web pages and continuously grab all web pages linked in the web page, the searching method is the best method for realizing a universal web crawler, and the method has the advantages of easy realization, prevention of trapping in an endless deep branch, parallel processing of the web crawler and improvement of the grabbing speed of the web crawler, and the analysis method of the crawler algorithm comprises the steps of popping up a reminding interface according to the access frequency of a user to the web pages, and ensuring stable signal transmission, Analyzing webpage data according to external links of webpages, levels of webpages, grades of webpages and the like, calculating weights of webpages, ranking the webpages, analyzing the webpage data according to content characteristics of the webpages, such as appearances, texts and the like, generating crawling instructions according to customer demands effectively to crawl needed contents, accelerating crawling speed and efficiency, checking crawling results of a crawler algorithm by a manual rechecking module, removing some illegal information, displaying structures, avoiding that all crawling results cannot be displayed after illegal contents appear, connecting office equipment, household equipment, cloud robots and production equipment with an internal database connected with a neural network system, collecting external signals through various settings, and collecting the collected signals regularly, A virtual system is created after storage and analysis, and signals are collected in multiple aspects, so that the diversity of sensory signals can be effectively guaranteed, and the accuracy of the signals in the system is guaranteed.
Drawings
FIG. 1 is a flow chart of a method of the present invention;
FIG. 2 is a block diagram of a neural network system of the present invention;
FIG. 3 is a partial method flow diagram of the present invention;
fig. 4 is a block diagram of the overall system connection of the present invention.
In the figure: 1. a neural network system; 11. a database interface module; 12. a crawler algorithm; 13. a calculation module; 14. crawling answer reference module; 15. a manual review module; 16. a display module; 2. a virtual system; 21. a virtual auditory system; 211. an audio acquisition module; 22. a virtual vision system; 221. a video acquisition module; 23. a virtual sensory system; 231. an air sensor; 232. a water system sensor; 233. a soil sensor; 24. a virtual motion system; 241. office equipment; 242. a household appliance; 243. a cloud robot; 244. production equipment; 3. a detection system; 31. a processing module; 311. and a signal sensing abnormity reminding module.
Detailed Description
The technical scheme in the embodiment of the invention will be made clear below by combining the attached drawings in the embodiment of the invention; fully described, it is to be understood that the described embodiments are merely exemplary of some, but not all, embodiments of the invention and that all other embodiments, which can be derived by one of ordinary skill in the art based on the described embodiments without inventive faculty, are within the scope of the invention.
Referring to fig. 1-2, the present invention provides the following technical solutions: an AI consultation based neural network building method comprises the following steps:
the method comprises the following steps: preparing a database, extracting characteristics as input to a neural network, calling and storing information content and collected content sent by an information initiator into an information content database by a database interface module 11 when the neural network is built;
step two: the method comprises the steps that a neural network system 1 is built, after the neural network system 1 is built, a crawler algorithm 12 in the system is connected with the Internet of things and the Internet through a database interface module 11, then crawler is conducted for multiple times, after calculation is conducted through the crawler algorithm 12, results are transmitted to a database through the database interface module 11 and stored;
step three: the neural network system 1 is arranged in an internet core server, the neural network system 1 is in butt joint with the virtual system 2 through a computing module 13, and the internet core server interacts with a demander through a human-computer interaction interface on a display module 16;
step four: the crawled answer quoting module 14 in the neural network system 1 quotes and feeds back the information and content crawled by the crawler algorithm 12, the information and content is displayed by the display module 16, and the verification and retrieval are carried out by the manual review module 15.
In the above embodiment, the database is used for correcting and storing a plurality of groups of crawled answers, the database interface module 11 is used for connecting the neural network system 1 with the internet of things and the internet, the detection system 3 for detecting the interface problem of the database interface module 11 is arranged in the neural network system 1, the processing module 31 for sensing whether the plurality of database interface modules 11 are connected with signals is arranged in the detection system 3, and the signal sensing abnormity prompting module 311 is arranged in the processing module 31, when a certain abnormal database interface module 11 is connected with a signal, the signal sensing abnormity prompting module 311 pops up a prompting interface to ensure stable transmission of the signal, wherein the signal sensing abnormity prompting module 311 can be used for sensing the plurality of database interface modules 11 to discharge the failed database interface module 11, so as to ensure stable transmission of the signal when data update information, the transmission efficiency is improved.
In the above embodiment, the web crawler of the crawler algorithm 12 in the neural network system 1 will access from a selected hyperlink according to a line and a link until reaching a leaf node of the line, that is, an HTML file without any hyperlink, and after processing the line, transfer to the next starting page, and continue to access one of the links included in the new starting page until reaching the leaf node, and then the web crawler will first capture all the web pages including the links in the starting page, then select one of the linked web pages, and continue to capture all the web pages linked in the web page, which is the best method for implementing a universal web crawler because it is characterized by easy implementation, and can avoid deep-layer trapping in an endless branch, and can make the web crawlers process in parallel, thereby improving the gripping speed thereof.
In the above embodiment, the webpage data crawled by the crawler node of the crawler algorithm 12 is stored in the resource library, the resource library analyzes the crawled data and establishes an index, the analysis method of the crawler algorithm 12 includes analyzing the webpage data according to the access frequency, the access duration, the click rate and the like of the user to the webpage, analyzing the webpage data according to the external link of the webpage, the hierarchy of the webpage and the like, calculating the weight of the webpage, ranking the webpage, analyzing the webpage data according to the content characteristics of the appearance of the webpage, the text of the webpage and the like, and generating a crawling instruction according to the client requirement effectively, so that the required content is crawled, and the crawling speed and efficiency are accelerated.
Please refer to fig. 3-4, an AI consultation-based neural network building system includes a neural network system 1 and a virtual system 2, the neural network system 1 is connected with the virtual system 2 through edge computing, wherein the neural network system 1 includes a database interface module 11, a crawler algorithm 12, a computing module 13, a crawl answer quoting module 14, an artificial rechecking module 15 and a display module 16, the database interface module 11 is connected with the neural network system 1, the crawler algorithm 12 is docked with the crawl answer quoting module 14 by a database, the computing module 13 is used for docking the virtual system 2, the artificial rechecking module 15 rechecks the crawl answer quoting module 14, the display module 16 can display the neural network system 1 on a terminal, a client designates a consultation answer database as required, deep learning, the crawler algorithm 12 in the neural network system 1 is in the designated consultation answer database according to a question, crawl the desired results and display the results on the display module 16;
the database interface module 11 is provided with a plurality of groups for the gap connection of different output ports, so as to ensure the stability of information transmission;
the manual review module 15 is used for labeling and classifying the data, multiple paths are used for reviewing the content, the result of the multiple paths is aggregated, the manual review module 15 is used for checking the result crawled by the crawler algorithm 12, after some illegal information is removed, the structure is displayed, after the illegal content is avoided, all crawled results cannot be displayed, and the crawled answer quoting module 14 is used for combing and quoting the information crawled by the crawler algorithm 12 for proofing.
In the above embodiment, the virtual system 2 includes a virtual auditory system 21, a virtual visual system 22, a virtual sensory system 23 and a virtual motion system 24, wherein the virtual auditory system 21, the virtual visual system 22, the virtual sensory system 23 and the virtual motion system 24 are all connected to the neural network system 1, the virtual auditory system 21 is composed of a plurality of audio acquisition modules 211, the virtual auditory system 21 stores and translates the audio content into text for later crawling by collecting the audio information acquired by the audio acquisition modules 211, the audio acquisition modules 211 are used for acquiring the voice data of the consultant and keeping better consistency on the extracted voice information, thereby reducing the influence probability of noise disturbance on the voice recognition rate, the virtual visual system 22 is composed of a plurality of video acquisition modules 221, the virtual visual system 22 acquires the video information, the video information is disassembled into audio content and frame-cutting content, then screening and checking are respectively carried out through an audio checking module and a picture checking module, the virtual sensing system 23 comprises an air sensor 231, a water system sensor 232 and a soil sensor 233, the air sensor 231, the water system sensor 232 and the soil sensor 233 are all connected with a collecting module in the virtual sensing system 23, collected sensing signals are transmitted and stored in the collecting module, the virtual motion system 24 comprises office equipment 241, household equipment 242, a cloud robot 243 and production equipment 244, the office equipment 241, the household equipment 242, the cloud robot 243 and the production equipment 244 are all connected with an internal database connected with the neural network system 1, external signals are collected through various settings, a virtual system 2 is created after the collected signals are structured, stored and analyzed, and the signals are collected through multiple aspects, the method can effectively ensure the diversity of sensory signals and ensure the accuracy of signals in the system.
In the above-described embodiment, the terminal for storing and installing the neural network system 1 and the display module 16 includes a mobile terminal such as a desktop, a collection, a notebook, and the like.
In conclusion; the neural network building method and system based on AI consultation ensure stable transmission of signals by popping up a reminding interface through the signal perception abnormity reminding module 311, wherein the signal perception abnormity reminding module 311 can be used for sensing a plurality of database interface modules 11 and eliminating a fault database interface module 11, so that the signals can be stably transmitted when data update information is obtained, the transmission efficiency is improved, a web crawler can first grab all web pages including links in an initial web page and then select one of the link web pages to continuously grab all web pages linked in the web page, the searching method is the best method for realizing a universal web crawler, and has the characteristics of easy realization, prevention of being trapped in an endless deep branch, parallel processing of the web crawler and improvement of the grabbing speed of the web crawler, and the analysis method of the crawler algorithm 12 comprises the steps of popping up a reminding interface according to the access frequency of a user to the web pages, and ensuring stable transmission of the signals, wherein the web crawler can grab all web pages including the initial web pages including links in the initial, Analyzing the webpage data according to the external link of the webpage, the hierarchy of the webpage, the grade of the webpage and the like, calculating the weight of the webpage, ranking the webpage, analyzing the webpage data according to the content characteristics of the appearance of the webpage, the text of the webpage and the like, effectively generating a crawling instruction according to the requirement of a client so as to crawl the required content and accelerate the crawling speed and efficiency, checking the crawling result of the crawler algorithm 12 by the manual rechecking module 15, displaying the structure after removing some illegal information, avoiding that all crawling results cannot be displayed after the illegal content appears, connecting the office equipment 241, the household equipment 242, the cloud robot 243 and the production equipment 244 with the internal database connected with the neural network system 1, and acquiring external signals through various settings, a virtual system 2 is created after the collected signals are structured, stored and analyzed, and the signals are collected in multiple aspects, so that the diversity of sense signals can be effectively guaranteed, and the accuracy of the signals in the system is guaranteed.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be able to cover the technical solutions and the inventive concepts of the present invention within the technical scope of the present invention.
Claims (9)
1. An AI consultation-based neural network building method is characterized by comprising the following steps of:
s1: preparing a database, extracting characteristics as input to a neural network, calling and storing information content and collected content sent by an information initiator into an information content database by a database interface module (11) when the neural network is built;
s2: the method comprises the steps that a neural network system (1) is built, after the neural network system (1) is built, a crawler algorithm (12) in the system is connected with the Internet of things and the Internet through a database interface module (11) for a plurality of times of crawlers, and after the crawler algorithm (12) is used for calculation, results are transmitted to a database through the database interface module (11) for storage;
s3: the neural network system (1) is arranged in the Internet core server, the neural network system (1) is in butt joint with the virtual system (2) through the computing module (13), and the Internet core server interacts with a demander through a human-computer interaction interface on the display module (16);
s4: a crawl answer quoting module (14) in the neural network system (1) quotes and feeds back information and contents crawled by the crawler algorithm (12), a display module (16) displays the information and the contents, and a manual review module (15) conducts verification and retrieval.
2. The AI-consultation-based neural network building method according to claim 1, wherein the database in S1 is used for correcting and storing a plurality of groups of crawled answers, the database interface module (11) is used for connecting the neural network system (1) with the Internet of things and the Internet, and the neural network system (1) is internally provided with a detection system (3) for detecting interface problems of the database interface module (11).
3. The AI-based consultation neural network building method according to claim 2, wherein a processing module (31) for sensing whether a plurality of database interface modules (11) are accessed to signals is arranged in the detection system (3), a signal sensing abnormity prompting module (311) is arranged in the processing module (31), and when a certain abnormal database interface module (11) is accessed to a signal, a prompting interface is popped up through the signal sensing abnormity prompting module (311) to ensure stable transmission of the signal; the signal perception abnormity reminding module (311) can be used for a plurality of database interface modules (11) to perceive and discharge fault database interface modules (11).
4. The AI-based consultation neural network building method of claim 3, wherein the web crawler of the crawling algorithm (12) in the neural network system (1) in S2 proceeds from a selected hyperlink to a line and a link from one link to another until reaching the leaf node of the line, i.e. the HTML file without any hyperlinks, and then proceeds to the next start page after processing the line, and continues to access one of the links included in the new start page until reaching the leaf node.
5. The AI-consultation-based neural network construction method of claim 4, wherein webpage data crawled by crawler nodes of the crawler algorithm (12) are stored in a resource library, and the resource library analyzes the crawled data and establishes an index.
6. The utility model provides a neural network system of buildding based on AI consultation which characterized in that: the system comprises a neural network system (1) and a virtual system (2), wherein the neural network system (1) is connected with the virtual system (2) through edge calculation; the neural network system (1) comprises a database interface module (11), a crawler algorithm (12), a calculation module (13), a crawl answer quoting module (14), an artificial rechecking module (15) and a display module (16), wherein the database interface module (11) is connected with the neural network system (1), the crawler algorithm (12) and the crawl answer quoting module (14) are butted through a database, the calculation module (13) is used for butting the virtual system (2), the artificial rechecking module (15) rechecks the crawl answer quoing module (14), and the display module (16) can display the neural network system (1) on a terminal;
the database interface module (11) is provided with a plurality of groups for the gap connection of different output ports, thereby ensuring the stability of information transmission;
the manual review module (15) performs labeling classification on the data, adopts a plurality of paths to review the content, and reviews the content result by aggregating the plurality of paths;
and the crawl answer quoting module (14) is used for combing and quoting evidence of the information crawled by the crawler algorithm (12).
7. The AI-based consultation cloud computing building system of claim 6, wherein: the virtual system (2) comprises a virtual auditory system (21), a virtual visual system (22), a virtual sensory system (23) and a virtual motion system (24), wherein the virtual auditory system (21), the virtual visual system (22), the virtual sensory system (23) and the virtual motion system (24) are all connected with the neural network system (1);
the virtual auditory system (21) is composed of a plurality of audio acquisition modules (211), and the virtual auditory system (21) stores and translates audio contents into characters for facilitating crawling in later period by collecting audio information acquired by the audio acquisition modules (211);
the virtual vision system (22) is composed of a plurality of video acquisition modules (221), the virtual vision system (22) acquires video information, disassembles the video information into audio content and frame-cutting content, and then screens and checks the audio content and the frame-cutting content through an audio checking module and a picture checking module respectively;
the virtual sensing system (23) comprises an air sensor (231), a water system sensor (232) and a soil sensor (233), wherein the air sensor (231), the water system sensor (232) and the soil sensor (233) are all connected with a collecting module in the virtual sensing system (23), and transmit and store the collected sensing signals into the collecting module.
8. The AI-based consultation cloud computing building system of claim 7, wherein: the virtual motion system (24) comprises office equipment (241), household equipment (242), a cloud robot (243) and production equipment (244), wherein the office equipment (241), the household equipment (242), the cloud robot (243) and the production equipment (244) are connected with the neural network system (1) connection internal database.
9. The AI-based consultation cloud computing building system of claim 8, wherein: the terminals for storing and installing the neural network system (1) and the display module (16) include mobile terminals such as desktop computers, collection computers, notebooks, and the like.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110459925.2A CN113076459A (en) | 2021-04-27 | 2021-04-27 | Neural network building method and system based on AI consultation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110459925.2A CN113076459A (en) | 2021-04-27 | 2021-04-27 | Neural network building method and system based on AI consultation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113076459A true CN113076459A (en) | 2021-07-06 |
Family
ID=76618790
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110459925.2A Pending CN113076459A (en) | 2021-04-27 | 2021-04-27 | Neural network building method and system based on AI consultation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113076459A (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102880607A (en) * | 2011-07-15 | 2013-01-16 | 舆情（香港）有限公司 | Dynamic network content grabbing method and dynamic network content crawler system |
| CN103823551A (en) * | 2013-03-17 | 2014-05-28 | 浙江大学 | System and method for realizing multidimensional perception of virtual interaction |
| CN105279272A (en) * | 2015-10-30 | 2016-01-27 | 南京未来网络产业创新有限公司 | Content aggregation method based on distributed web crawlers |
| CN105608134A (en) * | 2015-12-18 | 2016-05-25 | 盐城工学院 | Multithreading-based web crawler system and web crawling method thereof |
| CN107493353A (en) * | 2017-10-11 | 2017-12-19 | 宁波感微知著机器人科技有限公司 | A kind of intelligent robot cloud computing method based on contextual information |
| CN107633079A (en) * | 2017-09-25 | 2018-01-26 | 重庆邮电大学 | A kind of vehicle device natural language human-machine interactions algorithm based on database and neutral net |
| CN109710826A (en) * | 2018-11-29 | 2019-05-03 | 淮河水利委员会水文局(信息中心) | A kind of internet information artificial intelligence acquisition method and its system |
| WO2019174613A1 (en) * | 2018-03-14 | 2019-09-19 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for cloud computing |
| CN110286751A (en) * | 2019-02-26 | 2019-09-27 | 深圳市看见网络科技有限公司 | A kind of virtual scene digital applications exchange method |
| CN110489627A (en) * | 2019-08-12 | 2019-11-22 | 南京烽火天地通信科技有限公司 | A kind of internet crawler method for routing |
| CN111767443A (en) * | 2020-06-11 | 2020-10-13 | 西安石油大学 | Efficient web crawler analysis platform |
-
2021
- 2021-04-27 CN CN202110459925.2A patent/CN113076459A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102880607A (en) * | 2011-07-15 | 2013-01-16 | 舆情（香港）有限公司 | Dynamic network content grabbing method and dynamic network content crawler system |
| CN103823551A (en) * | 2013-03-17 | 2014-05-28 | 浙江大学 | System and method for realizing multidimensional perception of virtual interaction |
| CN105279272A (en) * | 2015-10-30 | 2016-01-27 | 南京未来网络产业创新有限公司 | Content aggregation method based on distributed web crawlers |
| CN105608134A (en) * | 2015-12-18 | 2016-05-25 | 盐城工学院 | Multithreading-based web crawler system and web crawling method thereof |
| CN107633079A (en) * | 2017-09-25 | 2018-01-26 | 重庆邮电大学 | A kind of vehicle device natural language human-machine interactions algorithm based on database and neutral net |
| CN107493353A (en) * | 2017-10-11 | 2017-12-19 | 宁波感微知著机器人科技有限公司 | A kind of intelligent robot cloud computing method based on contextual information |
| WO2019174613A1 (en) * | 2018-03-14 | 2019-09-19 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for cloud computing |
| CN109710826A (en) * | 2018-11-29 | 2019-05-03 | 淮河水利委员会水文局(信息中心) | A kind of internet information artificial intelligence acquisition method and its system |
| CN110286751A (en) * | 2019-02-26 | 2019-09-27 | 深圳市看见网络科技有限公司 | A kind of virtual scene digital applications exchange method |
| CN110489627A (en) * | 2019-08-12 | 2019-11-22 | 南京烽火天地通信科技有限公司 | A kind of internet crawler method for routing |
| CN111767443A (en) * | 2020-06-11 | 2020-10-13 | 西安石油大学 | Efficient web crawler analysis platform |
Non-Patent Citations (3)
| Title |
|---|
| 书虫子: "互联网进化论（2）", pages 1 - 13, Retrieved from the Internet <URL:https://m.sohu.com/a/339444544_488463?_trans_=010004_zyllqxtmrfx> * |
| 杜长燕;李祥龙;: "基于WEB的网络爬虫的设计", 无线互联科技, no. 05, 10 March 2015 (2015-03-10) * |
| 齐虎春: "大数据背景下的网络信息采集策略研究", 《内蒙古科技与经济》, no. 13, 15 July 2020 (2020-07-15), pages 79 - 81 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jauro et al. | Deep learning architectures in emerging cloud computing architectures: Recent development, challenges and next research trend | |
| US20190362219A1 (en) | Methods and systems for neural and cognitive processing | |
| Freire et al. | ManyNets: an interface for multiple network analysis and visualization | |
| CN110351244A (en) | A kind of network inbreak detection method and system based on multireel product neural network fusion | |
| CN110442790A (en) | Recommend method, apparatus, server and the storage medium of multi-medium data | |
| CN110490242A (en) | Training method, eye fundus image classification method and the relevant device of image classification network | |
| Kang et al. | Non-inertial opposition-based particle swarm optimization and its theoretical analysis for deep learning applications | |
| CN107169598A (en) | A kind of weather prediction method and system based on deep learning | |
| CN112491854B (en) | Multi-azimuth security intrusion detection method and system based on FCNN | |
| CN112087442B (en) | Time sequence related network intrusion detection method based on attention mechanism | |
| CN108664932A (en) | A kind of Latent abilities state identification method based on Multi-source Information Fusion | |
| Tan et al. | A new QoE-based prediction model for evaluating virtual education systems with COVID-19 side effects using data mining | |
| CN111625715B (en) | Information extraction method and device, electronic equipment and storage medium | |
| CN111782644A (en) | Scientific research data management control system and method based on block chain technology | |
| CN109934106A (en) | A kind of user behavior analysis method based on video image deep learning | |
| CN111563158A (en) | Text sorting method, sorting device, server and computer-readable storage medium | |
| CN114330966A (en) | Risk prediction method, device, equipment and readable storage medium | |
| CN112819024B (en) | Model processing method, user data processing method and device and computer equipment | |
| CN113076459A (en) | Neural network building method and system based on AI consultation | |
| Yang et al. | Multi-Head multimodal deep interest recommendation network | |
| Romney et al. | Curriculum for hands-on artificial intelligence cybersecurity | |
| Yifan | Application of machine learning in network security situational awareness | |
| Long et al. | 6G comprehensive intelligence: network operations and optimization based on Large Language Models | |
| Aktaş et al. | Sequential rule mining on the student behavior data of an e-learning platform in the field of financial sciences: Case study | |
| CN109635942B (en) | Brain excitation state and inhibition state imitation working state neural network circuit structure and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |