CN105528465A - Credit status assessment method and device - Google Patents
Credit status assessment method and device Download PDFInfo
- Publication number
- CN105528465A CN105528465A CN201610076337.XA CN201610076337A CN105528465A CN 105528465 A CN105528465 A CN 105528465A CN 201610076337 A CN201610076337 A CN 201610076337A CN 105528465 A CN105528465 A CN 105528465A
- Authority
- CN
- China
- Prior art keywords
- news
- factor
- favourable
- target subject
- data
- 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
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Abstract
An embodiment of the invention provides a credit status assessment method and device. The method comprises the following steps: all news data associated with a target body in a first preset time period are crawled in real time; each crawled news datum is subjected to word segmentation processing; a preset good news and bad news classification model is used for classifying the news data after word segmentation processing, obtained good news statuses or bad news statuses corresponding to each news datum are calculated, and a good news or bad news index of news associated with the target body is obtained; a plurality of current fixed income factors associated with the target body are acquired; the credit status of the target body is assessed with a preset credit assessment model according to the good news or bad news index of the news and the plurality of current fixed income factors.
Description
Technical field
The present invention relates to computer internet field, in particular to a kind of credit standing appraisal procedure and device.
Background technology
In prior art, utilize neural language program (NLP) technology can judge the positive and negative Sentiment orientation of internet public feelings, but do not have technical scheme by the investment logic associative system of internet public feelings and bond series products particularly fixed income series products, reason is that the positive adverse factor in the positive negative news of internet public feelings and investment logic exists larger difference.That is, the credit standing utilizing this entity of the assessment such as financial information, trade information, market information of the entity (i.e. company) of distribution fixed income series products, analyzes with auxiliary fixed income investment decision more at present.But credit standing that this method obtains is not enough comprehensively also not accurate enough.
Summary of the invention
Given this, the object of the present invention is to provide a kind of credit standing appraisal procedure and device, exist more comprehensively and more inaccurate problem the assessment of the credit standing of entity to improve in prior art.
To achieve these goals, the technical scheme of embodiment of the present invention employing is as follows:
First aspect, embodiments provides a kind of credit standing appraisal procedure, comprising: crawl all news data in the first preset time period be associated with target subject in real time; Word segmentation processing is carried out to crawled every bar news data; Utilize the empty profit disaggregated model favourable preset to carry out classification process to the every bar news data after word segmentation processing, the obtained favourable situation corresponding with described every bar news data or empty profit situation are added up to the message face advantage or empty profit index that obtain being associated with described target subject; Obtain the multiple current fixed income factor be associated with described target subject; Utilize the credit evaluation model preset, according to described message face, favourable or empty profit index and described multiple current fixed income factor assess the credit standing of described target subject.
Second aspect, the embodiment of the present invention additionally provides a kind of credit standing apparatus for evaluating, comprising: crawl module, for crawling all news data in the first preset time period be associated with target subject in real time; Word-dividing mode, for carrying out word segmentation processing to crawled every bar news data; Empty profit statistical module favourable, for utilizing default empty profit disaggregated model favourable to carry out classification process to the every bar news data after word segmentation processing, the obtained favourable situation corresponding with described every bar news data or empty profit situation are added up to the message face advantage or empty profit index that obtain being associated with described target subject; Acquisition module, for obtaining the multiple current fixed income factor be associated with described target subject; Evaluation module, for utilizing default credit evaluation model, according to described message face, favourable or empty profit index and described multiple current fixed income factor assess the credit standing of described target subject.
The credit standing appraisal procedure that the embodiment of the present invention provides and device, by crawling the news data be associated with target subject, to crawled every bar news data carry out word segmentation processing and empty profit favourable classification process obtain advantage corresponding to this news or empty profit situation, the favourable situation corresponding with every bar news data or empty profit situation are added up to the message face advantage or empty profit index that obtain being associated with described target subject, combine the credit standing that the fixed income factor integration be associated with target subject evaluates target subject afterwards.As can be seen here, in the scheme that the embodiment of the present invention provides, by internet public feelings being converted to the bearish factor favourable in investment logic, to take into full account may affecting of internet public feelings in the process of the credit standing of assessment objective main body, make it possible to the assessment accuracy of the credit standing improving target subject, contributing to promoting the auxiliary dynamics analyzed fixed income investment decision, is offering convenience property of user.
For making above and other object of the present invention, feature and advantage can become apparent, preferred embodiment cited below particularly, and coordinate institute's accompanying drawings, are described in detail below.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.Shown in accompanying drawing, above-mentioned and other object of the present invention, Characteristics and advantages will be more clear.Reference numeral identical in whole accompanying drawing indicates identical part.Deliberately do not draw accompanying drawing by physical size equal proportion convergent-divergent, focus on purport of the present invention is shown.
Fig. 1 is the applied environment schematic diagram of the embodiment of the present invention;
Fig. 2 shows a kind of structured flowchart that can be applicable to the server of the embodiment of the present invention;
Fig. 3 shows the process flow diagram of the credit standing appraisal procedure that first embodiment of the invention provides;
Fig. 4 shows the process flow diagram of the credit standing appraisal procedure that second embodiment of the invention provides;
Fig. 5 shows the schematic diagram of the credit standing apparatus for evaluating that third embodiment of the invention provides;
Fig. 6 shows the schematic diagram of the credit standing apparatus for evaluating that fourth embodiment of the invention provides.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, carry out clear, complete description to the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
It should be noted that: represent similar terms in similar label and letter accompanying drawing below, therefore, once be defined in an a certain Xiang Yi accompanying drawing, then do not need to define further it and explain in accompanying drawing subsequently.Meanwhile, in describing the invention, term " first ", " second " etc. only for distinguishing description, and can not be interpreted as instruction or hint relative importance.
The following each embodiment of the present invention all can be applicable in environment as shown in Figure 1.As shown in Figure 1, server 100 is communicated to connect by network 300 and one or more user terminal 200, to carry out data communication or mutual.Described server 100 can be multiple servers such as instant communication server, the webserver, database server, authentication server, also can be a server.Described user terminal 200 can be PC (personalcomputer, PC), panel computer, smart mobile phone, personal digital assistant (personaldigitalassistant, PDA) etc.User terminal 200 and server 100 can establish a communications link in Wi-Fi (Wireless Fidelity) network, 2G/3G/4G network or LAN (Local Area Network).
Fig. 2 shows a kind of structured flowchart of the server that can be applicable in the embodiment of the present invention.As shown in Figure 2, described server 100 can comprise credit standing apparatus for evaluating, storer 102, memory controller 103, processor 104 and the mixed-media network modules mixed-media 105 that the embodiment of the present invention provides.
Be electrically connected directly or indirectly between storer 102, memory controller 103, processor 104, each element of mixed-media network modules mixed-media 105, to realize the transmission or mutual of data.Such as, electrical connection can be realized by one or more communication bus or signal bus between these elements.Described credit standing apparatus for evaluating comprises at least one can be stored in software function module, the software function module that such as described credit standing apparatus for evaluating comprises or computer program in storer 102 with the form of software or firmware (firmware).
Storer 102 can store various software program and module, the credit standing appraisal procedure provided as the embodiment of the present invention and programmed instruction/module corresponding to device, processor 104 is by running the software program and module that store in the memory 102, thus perform the application of various function and data processing, namely realize the credit standing appraisal procedure in the embodiment of the present invention.Storer 102 can include but not limited to random access memory (RandomAccessMemory, RAM), ROM (read-only memory) (ReadOnlyMemory, ROM), programmable read only memory (ProgrammableRead-OnlyMemory, PROM), erasable read-only memory (ErasableProgrammableRead-OnlyMemory, EPROM), electricallyerasable ROM (EEROM) (ElectricErasableProgrammableRead-OnlyMemory, EEPROM) etc.
Processor 104 can be a kind of integrated circuit (IC) chip, has signal handling capacity.Above-mentioned processor can be general processor, comprises central processing unit (CentralProcessingUnit is called for short CPU), network processing unit (NetworkProcessor is called for short NP) etc.; Can also be digital signal processor (DSP), special IC (ASIC), ready-made programmable gate array (FPGA) or other programmable logic device (PLD), discrete gate or transistor logic, discrete hardware components.It can realize or perform disclosed each method, step and logic diagram in the embodiment of the present invention.The processor etc. of general processor can be microprocessor or this processor also can be any routine.
Mixed-media network modules mixed-media 105 is for receiving and sending network signal.Above-mentioned network signal can comprise wireless signal or wire signal.
Be appreciated that the structure shown in Fig. 2 is only signal, server 100 can also comprise than assembly more or less shown in Fig. 2, or has the configuration different from shown in Fig. 2.Each assembly shown in Fig. 2 can adopt hardware, software or its combination to realize.In addition, the server in the embodiment of the present invention can also comprise the server of multiple concrete difference in functionality.
In embodiments of the present invention, can be provided with client in user terminal 200, this client can be third-party application software, corresponding with server end, thus provides service for user, such as inquiry investment relevent information etc.The embodiment of the present invention is mainly based on the scene that server end utilizes the credit standing of news data and the multiple fixed income factor integration assessment objective main body be associated with target subject, by evaluated credit standing by client push to associated user, such as hold or pay close attention to the user of the fixed income product that this target subject is issued, so that user makes further investment decision accordingly.It should be noted that main body mentioned herein refers to the company of distribution fixed income series products.In addition, the scheme that the embodiment of the present invention provides preferably is applicable to fixed income series products investment analysis application scenarios.
First embodiment
Fig. 3 shows the process flow diagram of the credit standing appraisal procedure that first embodiment of the invention provides.Refer to Fig. 3, the credit standing appraisal procedure that the first embodiment provides can comprise:
Step S11, crawls all news data in the first preset time period be associated with target subject in real time.
Particularly, server real-time (such as every 5 minutes) can crawl all news data in the first preset time period be associated with described target subject from one or several the combination the portal website of each government organs website, financial web site, described target subject and social media website.The various policy data of the development trend affecting target subject, the assay data of mass media to target subject, the development strategy data etc. of target subject can be comprised in the news data obtained.Social media website can comprise microblogging, forum etc.In order to improve accuracy, should crawl all history news data be associated with described target subject as much as possible, that is, the first preset time period should be long as far as possible.Preferably, the first preset time period can be 2 years to 3 years.
Step S12, carries out word segmentation processing to crawled every bar news data.
After crawling the news data be associated with described target subject, server can carry out word segmentation processing to it, so that carry out subsequent arithmetic.Segmenting method well known in the prior art can be adopted to carry out word segmentation processing.Preferably, default participle model can be utilized to carry out word segmentation processing to crawled every bar news data.
Described default participle model can be set up in advance by following steps: the participle corpus obtaining basis, and based on described participle corpus, determine the observation state based on word, using the participle attribute of described word as hidden state (B/ prefix, in M/ word, E/ suffix, S/ individual character becomes word), transition matrix is built by the transition probability between described hidden state, determine the probability matrix of the initial time of described hidden state, and build confusion matrix with the interactive information (Gross entropy of joint distribution and product distribution) between the probability under described observation state hidden state described in each and described word, Laplce's smoothing processing is carried out to the parameter of the unlapped word of described participle corpus, using the Hidden Markov Model (HMM) that obtains thus as described default participle model.For the ease of understanding, suppose that participle expects that there are two words " A scientific & technical corporation has set up scientific and technological subsidiary company " and " B mechanism is on the verge of bankruptcy " in storehouse, observation state is " A, section, skill, public, department, , become, vertical, section, skill, son, public, department, B, machine, structure, be on the point of, face, broken, produce " these 20 words, hidden state comprises prefix (such as " section "), in word (such as " skill ", " public affairs "), suffix (such as " department "), individual character becomes word (such as " "), the probability that in transition matrix i.e. each hidden state, state transfers to another (such as, list becomes word to become the probability of word to list, list becomes word to the probability etc. of prefix) set, the probability distribution of initial time can be the probability (being 1/2) of A and B beginning in this instance, the probability that confusion matrix can show as observation state by hidden state is formed.
Step S13, utilize the empty profit disaggregated model favourable preset to carry out classification process to the every bar news data after word segmentation processing, the obtained favourable situation corresponding with described every bar news data or empty profit situation are added up to the message face advantage or empty profit index that obtain being associated with described target subject.
Because the positive negative news of internet public feelings exists larger difference with the positive adverse factor in investment logic, so need to carry out empty profit favourable classification process, the advantage corresponding to news data namely needing to determine that every bar crawls or empty profit degree to the news data crawled from internet.In a kind of embodiment, default empty profit disaggregated model favourable can be utilized to carry out classification process to the every bar news data after word segmentation processing.Utilize the empty profit disaggregated model favourable preset, every bar news data can be calculated respectively and belong to the probability of class favourable and belong to the probability of empty profit class, this news data is divided into the classification that greater probability in the probability belonging to calculated is corresponding.
Such as, if calculate the probability that news item data belong to class favourable to be greater than its probability belonging to empty profit class, can determine that this news data belongs to class favourable.Similarly, if calculate the probability that news item data belong to class favourable to be less than its probability belonging to empty profit class, can determine that this news data belongs to empty profit class.After calculating situation favourable corresponding to every bar news data or empty profit situation, usually first can carry out duplicate removal process for all news data, afterwards statistical treatment be carried out to the news data after duplicate removal.Preferably, the ratio that the number being divided into the news data of empty profit class in all news data in the first crawled preset time period accounts for the total number of described all news data can be calculated, the favourable or empty profit index as the message face be associated with described target subject.But, the ratio that the number being divided into the news data of class favourable in all news data in the first crawled preset time period accounts for the total number of described all news data can also be calculated, the favourable or empty profit index as the message face be associated with described target subject.In addition, can also adopt other statisticals, the specific embodiment of the invention is not as limit.
In a kind of embodiment, described default empty profit disaggregated model favourable can be set up in advance by following steps: obtain the bad news sample multiple favourable for fixed income business, each described bad news sample favourable comprise one participle with the first history news data that a main body is associated and correspondingly with the first history news data of this participle determine advantage or empty profit situation; One group is divided into by described multiple bad news sample favourable to train with bad news sample favourable and one group of test bad news sample favourable; Utilize described one group of training with bad news sample favourable to Bayesian model or the training of adaboosting model, utilize described one group of test to test to housebroken Bayesian model or adaboosting model with bad news sample favourable, and assay meet first pre-conditioned time be described default empty profit disaggregated model favourable by described housebroken Bayesian model or adaboosting model specification.
Set up in the process of empty profit disaggregated model favourable in advance above-mentioned, described multiple bad news sample favourable can be that fixed income researcher obtains for the arrangement of fixed income business.The bad news sample multiple favourable obtained can be associated with multiple main body respectively, and some in wherein said multiple bad news sample favourable can also be associated with same main body.Each the first history news data of participle correspond to situation favourable or empty profit situation and can be determined by fixed income researcher.
When number relatively less (being small data set) of bad news sample favourable, server can be trained Bayesian model (speed of convergence is very fast), correspondingly assay meet first pre-conditioned time housebroken Bayesian model is set as described default empty profit disaggregated model favourable.When number relatively many (being large data sets) of bad news sample favourable, can to the training of adaboosting model, correspondingly assay meet first pre-conditioned time be described default empty profit disaggregated model favourable by housebroken adaboosting model specification.Assay meets first and pre-conditionedly can to represent: utilize described one group of test bad news sample favourable to exceed predetermined threshold value to housebroken Bayesian model or the adaboosting model accuracy obtained of testing, this predetermined threshold value can set as the case may be, and such as 85% or 90%.
In addition, can also set up by the keyword extracted in above-mentioned multiple advantage and bad news sample the empty profit disaggregated model favourable preset.That is, empty profit classification favourable can be carried out by the method for Keywords matching to every bar news data.
Step S14, obtains the multiple current fixed income factor be associated with described target subject.
In a kind of embodiment, the multiple current fixed income factor that described acquisition is associated with described target subject can comprise: obtain be associated with described target subject the current debt paying ability factor, the operation ability factor, the growing ability factor, the earning capacity factor, the cash flow factor, the asset-liability ratio factor, the industry factor, in the produce market Summing Factor operation control factor at least partially.
The described debt paying ability factor can comprise asset-liability ratio (total liabilities/total assets), liquidity rate (floating assets/current liability), current rate ((floating assets-stock)/current liability), interest cover ratio (IBTT/interest expense), tangible net worth debt ratio (total liabilities/(stockholder's equity-intangible asset value)), cash ratio (cash flow/current liability from managing), equity radio (stockholder's equity/total assets) and fixed assets ratio (capital assets/total assets).
The described operation ability factor can comprise the turnover of total assets (sales revenue/average total assets), the velocity of liquid assets (sales revenue/floating assets average balance), stock turnover rate (selling cost/average inventory), accounts receivable turnover (sales revenue/average accounts receivable) and stockholder's equity turnover rate (sales revenue/stockholder's equity).
The described growing ability factor can comprise accumulation of capital rate (this year owner's equity troughput/beginning of the year owner's equity), profit after tax rate of growth (profit after tax (this year profit after tax-last year profit after tax)/last year) and total assets rate of growth (total assets in (this year total assets-last year total assets)/this year).
The described earning capacity factor can comprise net assets income ratio (net profit/average net assets), income from main operation rate ((main business income-main business cost)/main business income), return on total assets (net profit/average total assets) and net profit on sales rate (net profit/sales revenue).
The described cash flow factor comprises cash flow debt ratio (cash flow/current liability) and per share management road (cash net inflow/average shareholding equity).
The described asset-liability ratio factor comprises the net weight ratio (current liability/assets) of asset size (Ln (assets)) and current liability and assets.
The described industry factor comprises cost structure, industry cycle, industry profit, industry maturity and position in industry.The described produce market factor comprises product quality, product substitution, properties of product, market competition and supply, production and sale fluency.The described operation control factor comprises management and control ability, administration structure science degree, administration and supervision authorities stability, administration and supervision authorities adaptability to changes, cohesiveness, corporate culture, information weight * number, opportunistic practice in violation of rules and regulations.
Described multiple current fixed income factor can obtain from presetting database, and the operation control factor described in the wherein said industry factor, described produce market Summing Factor is normally drawn by specialty analysis (such as professional and technical personnel and/or related application analysis).Described presetting database can be built in described server, also can be placed on described server.
Step S15, utilizes the credit evaluation model preset, and according to described message face, favourable or empty profit index and described multiple current fixed income factor assess the credit standing of described target subject.
After obtaining the message face that is associated with described target subject index favourable or empty profit index and multiple current fixed income factor, using obtained all data as the credit standing that the input of default credit evaluation model can evaluate described target subject be excellent, good, in, poor or poor.
In a kind of embodiment, described default credit evaluation model can be set up in advance by following steps: obtain many historical sample data, historical sample data described in every bar is corresponding with a sample main body, and comprise the message face history advantage or empty profit index that are associated with this sample main body, multiple history fixed income factor, and history real credit situation, the described history fixed income factor is that described history real credit situation to be fixed time some a fixed income factor corresponding to place really, wherein said history message face advantage or empty profit index are that all second history news data in the second preset time period by crawling history carry out word segmentation processing, described default empty profit disaggregated model favourable is utilized to obtain carry out the classifying process adding up sorted result of every bar second history news data through word segmentation processing, described in every bar, the second history news data is all associated with described sample main body, the stop time point of described second preset time period early than and be close to described history real credit situation and really fix time a little, described many historical sample data are divided into one group of training sample data and one group of test sample book data, described one group of training sample data are utilized to train (i.e. each parameter of neural network training model) neural network model, utilize described one group of housebroken neural network model of test sample book data detection, and assay meet second pre-conditioned time described housebroken neural network model is set as described default credit evaluation model.
Build one's credit in advance in the process of assessment models above-mentioned, some in many historical sample data obtained can be associated with same sample main body.Described second preset time period preferably equals described first preset time period.Preferably, described one group of training sample data can account for 70% of described many historical sample data, described one group of test sample book data can account for 30% of described many historical sample data, but above-mentioned number percent can also change according to actual conditions, but should guarantee that the number of training sample data is greater than the number of test sample book data.Assay meets second and pre-conditionedly can to represent: the accuracy utilizing described one group of housebroken neural network model of test sample book data detection to obtain exceedes predetermined threshold value, and this predetermined threshold value can set as the case may be, and such as 90%.
Multiple history fixed income factors that every bar historical sample data comprises can comprise the debt paying ability factor, the operation ability factor, the growing ability factor, the earning capacity factor, the cash flow factor, the asset-liability ratio factor, the industry factor, the produce market Summing Factor operation control factor of the history be associated with the sample main body corresponding to this historical sample data.Detailed description about above-mentioned each factor refers to above.
The credit standing appraisal procedure that first embodiment of the invention provides, by crawling the news data be associated with target subject, to crawled every bar news data carry out word segmentation processing and empty profit favourable classification process obtain advantage corresponding to this news or empty profit situation, the favourable situation corresponding with every bar news data or empty profit situation are added up to the message face advantage or empty profit index that obtain being associated with described target subject, combine the credit standing that the fixed income factor integration be associated with target subject evaluates target subject afterwards.As can be seen here, in the scheme that the embodiment of the present invention provides, by internet public feelings being converted to the bearish factor favourable in investment logic, to take into full account may affecting of internet public feelings in the process of the credit standing of assessment objective main body, make it possible to the assessment accuracy of the credit standing improving target subject, contributing to promoting the auxiliary dynamics analyzed fixed income investment decision, is offering convenience property of user.
Second embodiment
Fig. 4 shows the process flow diagram of the credit standing evaluation method that second embodiment of the invention provides.Refer to Fig. 4, the difference of the credit standing appraisal procedure that second embodiment of the invention provides and the credit standing appraisal procedure that first embodiment of the invention provides mainly is: the step S16 performed after the credit standing appraisal procedure that the second embodiment provides can also be included in step S12 and the step S17 performed after step S15.The second embodiment part that is basically the same as those in the first embodiment refers to the associated description of the first embodiment, no longer repeats herein.
Step S16, by the data match of the critical event in every bar news data and predeterminable event storehouse of word segmentation processing, whenever the data match of a news data through word segmentation processing and a described critical event, push the notice corresponding with this critical event to the user holding the fixed income series products issued by described target subject.It should be noted that step S16 can perform with the arbitrary step in step S13, S14, S15 simultaneously, or can perform after step S15.
Particularly, the critical event in described predeterminable event storehouse can be the critical event of history precipitation, comprise can obviously the operation of target subject described in positive influences event with can the event of operation of target subject described in obvious negative effect.The data of critical event can be the key word of critical event.Such as, for a news data through word segmentation processing, when to match with the data of a described critical event as " governmental support " from it, the critical event of this news data with " item to manage of target subject obtains government and supports energetically " can be associated, and therefore to holding user's sending out notice of the fixed income series products issued by described target subject to inform the above-mentioned critical event of this user.Above-mentioned coupling can be performed by WM algorithm, AC algorithm or other suitable algorithms.In this way, the news data that can be associated with target subject in conjunction with predeterminable event storehouse Real-Time Monitoring, thus realize the real-time early warning to the major event of target subject, make investment decision with assisted user.
Step S17, pushes to the user holding the fixed income series products issued by described target subject by the credit standing of described target subject.
After obtaining the credit standing of described target subject by described default credit evaluation model assessment, server can be in several ways, such as client, Email, short message, webpage etc., the credit standing of target subject is pushed to user, with the concrete credit standing making user learn target subject in time.User can make investment decision according to learnt credit standing, such as target subject credit standing as well time continue to hold the fixed income series products issued by targeted company, and the credit standing of target subject be difference time select abandon holding the fixed income series products issued by targeted company.
Preferably, in step S17, except push the credit standing of described target subject to described user except, news data up-to-date in crawled news data can also be pushed to described user, can know that to make described user the news of described target subject is dynamic, contributes to described user and makes fixed investment decision-making in time.
The technical scheme that second embodiment provides is except all technique effects of technical scheme that can realize the first embodiment and provide, can also by realizing the real-time early warning to the major event of target subject in conjunction with predeterminable event storehouse, interested for user its information making fixed income investment decision that contributes to can be sent to user in addition, be offering convenience property of user.
3rd embodiment
Third embodiment of the invention provides a kind of credit standing apparatus for evaluating.Refer to Fig. 5, the credit standing apparatus for evaluating 300A that third embodiment of the invention provides can comprise:
Crawl module 310, for crawling all news data in the first preset time period be associated with target subject in real time;
Word-dividing mode 320, for carrying out word segmentation processing to crawled every bar news data;
Empty profit statistical module 330 favourable, for utilizing default empty profit disaggregated model favourable to carry out classification process to the every bar news data after word segmentation processing, the obtained favourable situation corresponding with described every bar news data or empty profit situation are added up to the message face advantage or empty profit index that obtain being associated with described target subject;
Acquisition module 340, for obtaining the multiple current fixed income factor be associated with described target subject;
Evaluation module 350, for utilizing default credit evaluation model, according to described message face, favourable or empty profit index and described multiple current fixed income factor assess the credit standing of described target subject.
In a kind of embodiment, described in crawl module 310 and from one or several the combination the portal website of each government organs website, financial web site, described target subject and social media website, real-time (such as every 5 minutes) all news data in the first preset time period be associated with described target subject can be crawled.
In a kind of embodiment, described word-dividing mode 320 can utilize default participle model to carry out word segmentation processing to crawled every bar news data.Further, consult Fig. 5, credit standing apparatus for evaluating 300A can also comprise participle model presetting module 360.
Described participle model presetting module 360 may be used for: the participle corpus obtaining basis, and based on described participle corpus, determine the observation state based on word, using the participle attribute of described word as hidden state (B/ prefix, in M/ word, E/ suffix, S/ individual character becomes word), transition matrix is built by the transition probability between described hidden state, determine the probability matrix of the initial time of described hidden state, and build confusion matrix with the interactive information (Gross entropy of joint distribution and product distribution) between the probability under described observation state hidden state described in each and described word, Laplce's smoothing processing is carried out to the parameter of the unlapped word of described participle corpus, using the Hidden Markov Model (HMM) that obtains thus as described default participle model.
Further, consult Fig. 5, credit standing apparatus for evaluating 300A can also comprise disaggregated model presetting module 370.Described disaggregated model presetting module 370 may be used for: obtain the bad news sample multiple favourable for fixed income business, each described bad news sample favourable comprise one participle with the first history news data that a main body is associated and correspondingly with the first history news data of this participle determine advantage or empty profit situation; One group is divided into by described multiple bad news sample favourable to train with bad news sample favourable and one group of test bad news sample favourable; Utilize described one group of training with bad news sample favourable to Bayesian model or the training of adaboosting model, utilize described one group of test to test to housebroken Bayesian model or adaboosting model with bad news sample favourable, and assay meet first pre-conditioned time be described default empty profit disaggregated model favourable by described housebroken Bayesian model or adaboosting model specification.
In a kind of embodiment, the multiple current fixed income factor be associated with described target subject that described acquisition module 340 obtains can comprise: in the current debt paying ability factor be associated with described target subject, the operation ability factor, the growing ability factor, the earning capacity factor, the cash flow factor, the asset-liability ratio factor, the industry factor, the produce market Summing Factor operation control factor at least partially.
Further, consult Fig. 5, credit standing apparatus for evaluating 300A can also comprise credit evaluation model presetting module 380.Described credit evaluation model presetting module 380 may be used for: obtain many historical sample data, historical sample data described in every bar is corresponding with a sample main body, and comprise the history message face advantage or empty profit index that are associated with this sample main body, multiple history fixed income factor, and history real credit situation, the described history fixed income factor is that described history real credit situation to be fixed time some a fixed income factor corresponding to place really, wherein said history message face advantage or empty profit index are that all second history news data in the second preset time period by crawling history carry out word segmentation processing, described default empty profit disaggregated model favourable is utilized to obtain carry out the classifying process adding up sorted result of every bar second history news data through word segmentation processing, described in every bar, the second history news data is all associated with described sample main body, the stop time point of described second preset time period early than and be close to described history real credit situation and really fix time a little, described many historical sample data are divided into one group of training sample data and one group of test sample book data, described one group of training sample data are utilized to train (i.e. each parameter of neural network training model) neural network model, utilize described one group of housebroken neural network model of test sample book data detection, and assay meet second pre-conditioned time described housebroken neural network model is set as described default credit evaluation model.
The present embodiment, to the detailed process of each Implement of Function Module of credit standing apparatus for evaluating 300A function separately, refers to the particular content of above-mentioned Fig. 1 to middle description embodiment illustrated in fig. 3, repeats no more herein.
The credit standing apparatus for evaluating that third embodiment of the invention provides, by crawling the news data be associated with target subject, to crawled every bar news data carry out word segmentation processing and empty profit favourable classification process obtain advantage corresponding to this news or empty profit situation, the favourable situation corresponding with every bar news data or empty profit situation are added up to the message face advantage or empty profit index that obtain being associated with described target subject, combine the credit standing that the fixed income factor integration be associated with target subject evaluates target subject afterwards.As can be seen here, in the scheme that the embodiment of the present invention provides, by internet public feelings being converted to the bearish factor favourable in investment logic, to take into full account may affecting of internet public feelings in the process of the credit standing of assessment objective main body, make it possible to the assessment accuracy of the credit standing improving target subject, contributing to promoting the auxiliary dynamics analyzed fixed income investment decision, is offering convenience property of user.
4th embodiment
Fig. 6 shows the schematic diagram of the credit standing evaluating apparatus that fourth embodiment of the invention provides.Refer to Fig. 6, the difference of the credit standing apparatus for evaluating 300A that the credit standing apparatus for evaluating 300B that fourth embodiment of the invention provides and third embodiment of the invention provide mainly is: credit standing apparatus for evaluating 300B can also comprise matching module 390 and pushing module 400.
Described matching module 390 may be used for: by the data match of the critical event in every bar news data and predeterminable event storehouse of word segmentation processing, whenever the data match of a news data through word segmentation processing and a described critical event, push the notice corresponding with this critical event to the user holding the fixed income series products issued by described target subject.
Described pushing module 400 may be used for: after described evaluation module 350 assessment obtains the credit standing of described target subject, the credit standing of described target subject is pushed to the user holding the fixed income series products issued by described target subject.
Further, described pushing module 400 can also be used for: while pushing in the described credit standing by described target subject the user holding the fixed income series products issued by described target subject, news data up-to-date in crawled news data is pushed to described user.
The present embodiment, to the detailed process of each Implement of Function Module of credit standing apparatus for evaluating 300B function separately, refers to the particular content of above-mentioned Fig. 1 to middle description embodiment illustrated in fig. 4, repeats no more herein.
It should be noted that, each embodiment in this instructions all adopts the mode of going forward one by one to describe, and what each embodiment stressed is the difference with other embodiments, between each embodiment identical similar part mutually see.For device class embodiment, due to itself and embodiment of the method basic simlarity, so description is fairly simple, relevant part illustrates see the part of embodiment of the method.
In several embodiments that the application provides, should be understood that disclosed apparatus and method also can realize by another way.Device embodiment described above is only schematic, and such as, the process flow diagram in accompanying drawing and block diagram show device according to multiple embodiment of the present invention, the architectural framework in the cards of method and computer program product, function and operation.In this, each square frame in process flow diagram or block diagram can represent a part for module, program segment or a code, and a part for described module, program segment or code comprises one or more executable instruction for realizing the logic function specified.Also it should be noted that at some as in the implementation of replacing, the function marked in square frame also can be different from occurring in sequence of marking in accompanying drawing.Such as, in fact two continuous print square frames can perform substantially concurrently, and they also can perform by contrary order sometimes, and this determines according to involved function.Also it should be noted that, the combination of the square frame in each square frame in block diagram and/or process flow diagram and block diagram and/or process flow diagram, can realize by the special hardware based system of the function put rules into practice or action, or can realize with the combination of specialized hardware and computer instruction.
In addition, each functional module in each embodiment of the present invention can integrate formation one independently part, also can be modules individualism, also can form an independently part by two or more module integrations.
If described function using the form of software function module realize and as independently production marketing or use time, can be stored in a computer read/write memory medium.Based on such understanding, the part of the part that technical scheme of the present invention contributes to prior art in essence in other words or this technical scheme can embody with the form of software product, this computer software product is stored in a storage medium, comprising some instructions in order to make a computer equipment (can be personal computer, server, or the network equipment etc.) perform all or part of step of method described in each embodiment of the present invention.And aforesaid storage medium comprises: USB flash disk, portable hard drive, ROM (read-only memory) (ROM, Read-OnlyMemory), random access memory (RAM, RandomAccessMemory), magnetic disc or CD etc. various can be program code stored medium.It should be noted that, in this article, the such as relational terms of first and second grades and so on is only used for an entity or operation to separate with another entity or operational zone, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or equipment.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment comprising described key element and also there is other identical element.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.It should be noted that: represent similar terms in similar label and letter accompanying drawing below, therefore, once be defined in an a certain Xiang Yi accompanying drawing, then do not need to define further it and explain in accompanying drawing subsequently.
The above; be only the specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, is anyly familiar with those skilled in the art in the technical scope that the present invention discloses; change can be expected easily or replace, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should described be as the criterion with the protection domain of claim.
Claims (10)
1. a credit standing appraisal procedure, is characterized in that, comprising:
Crawl all news data in the first preset time period be associated with target subject in real time;
Word segmentation processing is carried out to crawled every bar news data;
Utilize the empty profit disaggregated model favourable preset to carry out classification process to the every bar news data after word segmentation processing, the obtained favourable situation corresponding with described every bar news data or empty profit situation are added up to the message face advantage or empty profit index that obtain being associated with described target subject;
Obtain the multiple current fixed income factor be associated with described target subject;
Utilize the credit evaluation model preset, according to described message face, favourable or empty profit index and described multiple current fixed income factor assess the credit standing of described target subject.
2. method according to claim 1, is characterized in that, after carrying out word segmentation processing to crawled every bar news data, described method also comprises:
By the data match of the critical event in every bar news data and predeterminable event storehouse of word segmentation processing, whenever the data match of a news data through word segmentation processing and a described critical event, push the notice corresponding with this critical event to the user holding the fixed income series products issued by described target subject.
3. method according to claim 1, is characterized in that, after assessment obtains the credit standing of described target subject, described method also comprises:
The credit standing of described target subject is pushed to the user holding the fixed income series products issued by described target subject.
4. method according to claim 3, it is characterized in that, while pushing in the described credit standing by described target subject the user holding the fixed income series products issued by described target subject, news data up-to-date in crawled news data is pushed to described user.
5. method according to claim 1, is characterized in that, describedly crawls all news data in the first preset time period be associated with described target subject in real time, comprising:
The all news data in the first preset time period be associated with described target subject are crawled in real time in one or several combination from the portal website and social media website of each government organs website, financial web site, described target subject.
6. method according to claim 1, is characterized in that, described default empty profit disaggregated model favourable is set up in advance by following steps:
Obtain for the bad news sample multiple favourable of fixed income business, each described bad news sample favourable comprise one participle with the first history news data that a main body is associated and correspondingly with the first history news data of this participle determine advantage or empty profit situation;
One group is divided into by described multiple bad news sample favourable to train with bad news sample favourable and one group of test bad news sample favourable;
Utilize described one group of training with bad news sample favourable to Bayesian model or the training of adaboosting model, utilize described one group of test to test to housebroken Bayesian model or adaboosting model with bad news sample favourable, and assay meet first pre-conditioned time be described default empty profit disaggregated model favourable by described housebroken Bayesian model or adaboosting model specification.
7. method according to claim 1, is characterized in that, described default credit evaluation model is set up in advance by following steps:
Obtain many historical sample data, historical sample data described in every bar is corresponding with a sample main body, and comprise the history message face advantage or empty profit index, multiple history fixed income factor and history real credit situation that are associated with this sample main body, the described history fixed income factor is that described history real credit situation to be fixed time some a fixed income factor corresponding to place really
Wherein said history message face advantage or empty profit index are that all second history news data in the second preset time period by crawling history are carried out word segmentation processing, utilized described default empty profit disaggregated model favourable to carry out classification process and add up sorted result obtaining to the every bar second history news data through word segmentation processing, described in every bar, the second history news data is all associated with described sample main body, the stop time point of described second preset time period early than and be close to described history real credit situation and really fix time a little
Described many historical sample data are divided into one group of training sample data and one group of test sample book data;
Described one group of training sample data are utilized to train neural network model, utilize described one group of housebroken neural network model of test sample book data detection, and assay meet second pre-conditioned time described housebroken neural network model is set as described default credit evaluation model.
8. method according to claim 1, is characterized in that, utilize the participle model preset to carry out word segmentation processing to crawled every bar news data, described default participle model is set up in advance by following steps:
Obtain the participle corpus on basis, and based on described participle corpus, determine the observation state based on word, using the participle attribute of described word as hidden state, transition matrix is built by the transition probability between described hidden state, determine the probability matrix of the initial time of described hidden state, and build confusion matrix with the interactive information between the probability under described observation state hidden state described in each and described word;
Laplce's smoothing processing is carried out to the parameter of the unlapped word of described participle corpus, using the Hidden Markov Model (HMM) that obtains thus as described default participle model.
9. method according to claim 1, is characterized in that, the multiple current fixed income factor that described acquisition is associated with described target subject, comprising:
In the current debt paying ability factor that acquisition is associated with described target subject, the operation ability factor, the growing ability factor, the earning capacity factor, the cash flow factor, the asset-liability ratio factor, the industry factor, the produce market Summing Factor operation control factor at least partially
The described debt paying ability factor comprises asset-liability ratio, liquidity rate, current rate, interest cover ratio, tangible net worth debt ratio, cash ratio, equity radio and fixed assets ratio,
The described operation ability factor comprises the turnover of total assets, the velocity of liquid assets, stock turnover rate, accounts receivable turnover and stockholder's equity turnover rate,
The described growing ability factor comprises accumulation of capital rate, profit after tax rate of growth and total assets rate of growth,
The described earning capacity factor comprises net assets income ratio, income from main operation rate, return on total assets and net profit on sales rate,
The described cash flow factor comprises cash flow debt ratio and per share management road,
The described asset-liability ratio factor comprises the net weight ratio of asset size and current liability and assets,
The described industry factor comprises cost structure, industry cycle, industry profit, industry maturity and position in industry,
The described produce market factor comprises product quality, product substitution, properties of product, market competition and supply, production and sale fluency,
The described operation control factor comprises management and control ability, administration structure science degree, administration and supervision authorities stability, administration and supervision authorities adaptability to changes, cohesiveness, corporate culture, information weight * number, opportunistic practice in violation of rules and regulations.
10. a credit standing apparatus for evaluating, is characterized in that, comprising:
Crawl module, for crawling all news data in the first preset time period be associated with target subject in real time;
Word-dividing mode, for carrying out word segmentation processing to crawled every bar news data;
Empty profit statistical module favourable, for utilizing default empty profit disaggregated model favourable to carry out classification process to the every bar news data after word segmentation processing, the obtained favourable situation corresponding with described every bar news data or empty profit situation are added up to the message face advantage or empty profit index that obtain being associated with described target subject;
Acquisition module, for obtaining the multiple current fixed income factor be associated with described target subject;
Evaluation module, for utilizing default credit evaluation model, according to described message face, favourable or empty profit index and described multiple current fixed income factor assess the credit standing of described target subject.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610076337.XA CN105528465A (en) | 2016-02-03 | 2016-02-03 | Credit status assessment method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610076337.XA CN105528465A (en) | 2016-02-03 | 2016-02-03 | Credit status assessment method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105528465A true CN105528465A (en) | 2016-04-27 |
Family
ID=55770688
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610076337.XA Pending CN105528465A (en) | 2016-02-03 | 2016-02-03 | Credit status assessment method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105528465A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107481132A (en) * | 2017-08-02 | 2017-12-15 | 上海前隆信息科技有限公司 | A kind of credit estimation method and system, storage medium and terminal device |
| CN107563588A (en) * | 2017-07-25 | 2018-01-09 | 北京拓明科技有限公司 | A kind of acquisition methods of personal credit and acquisition system |
| CN107909274A (en) * | 2017-11-17 | 2018-04-13 | 平安科技（深圳）有限公司 | Enterprise investment methods of risk assessment, device and storage medium |
| CN108038103A (en) * | 2017-12-18 | 2018-05-15 | 北京百分点信息科技有限公司 | A kind of method, apparatus segmented to text sequence and electronic equipment |
| WO2018157808A1 (en) * | 2017-03-01 | 2018-09-07 | 腾讯科技（深圳）有限公司 | Method and system for evaluating internet credit |
| WO2018188543A1 (en) * | 2017-04-14 | 2018-10-18 | 腾讯科技（深圳）有限公司 | Real-time credit score adjustment processing method and device and processing server |
| CN109146611A (en) * | 2018-07-16 | 2019-01-04 | 浙江大学 | A kind of electric business product quality credit index analysis method and system |
| CN110782332A (en) * | 2019-01-18 | 2020-02-11 | 搜信信用产业集团有限公司 | Intelligent credit assessment dynamic tracing method |
| CN111125561A (en) * | 2019-11-28 | 2020-05-08 | 泰康保险集团股份有限公司 | Network heat display method and device |
| CN112148764A (en) * | 2019-06-28 | 2020-12-29 | 北京百度网讯科技有限公司 | Feature screening method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101055639A (en) * | 2006-04-14 | 2007-10-17 | 武汉群胜科技有限公司 | Bank credit operation processing system |
| CN101493913A (en) * | 2008-01-23 | 2009-07-29 | 阿里巴巴集团控股有限公司 | Method and system for assessing user credit in internet |
| CN102663650A (en) * | 2012-03-14 | 2012-09-12 | 钟文清 | System for analyzing enterprise credit risk and application method thereof |
| CN105022825A (en) * | 2015-07-22 | 2015-11-04 | 中国人民解放军国防科学技术大学 | Financial variety price prediction method capable of combining financial news mining and financial historical data |
| CN105260933A (en) * | 2015-11-10 | 2016-01-20 | 魏楠 | Loan appraisal method for inclusive finance |
-
2016
- 2016-02-03 CN CN201610076337.XA patent/CN105528465A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101055639A (en) * | 2006-04-14 | 2007-10-17 | 武汉群胜科技有限公司 | Bank credit operation processing system |
| CN101493913A (en) * | 2008-01-23 | 2009-07-29 | 阿里巴巴集团控股有限公司 | Method and system for assessing user credit in internet |
| CN102663650A (en) * | 2012-03-14 | 2012-09-12 | 钟文清 | System for analyzing enterprise credit risk and application method thereof |
| CN105022825A (en) * | 2015-07-22 | 2015-11-04 | 中国人民解放军国防科学技术大学 | Financial variety price prediction method capable of combining financial news mining and financial historical data |
| CN105260933A (en) * | 2015-11-10 | 2016-01-20 | 魏楠 | Loan appraisal method for inclusive finance |
Non-Patent Citations (1)
| Title |
|---|
| 朱双东: "《神经网络应用基础》", 31 March 2000, 东北大学出版社 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018157808A1 (en) * | 2017-03-01 | 2018-09-07 | 腾讯科技（深圳）有限公司 | Method and system for evaluating internet credit |
| CN108537397A (en) * | 2017-03-01 | 2018-09-14 | 腾讯科技（深圳）有限公司 | A kind of internet reference appraisal procedure and system |
| WO2018188543A1 (en) * | 2017-04-14 | 2018-10-18 | 腾讯科技（深圳）有限公司 | Real-time credit score adjustment processing method and device and processing server |
| CN107563588A (en) * | 2017-07-25 | 2018-01-09 | 北京拓明科技有限公司 | A kind of acquisition methods of personal credit and acquisition system |
| CN107481132A (en) * | 2017-08-02 | 2017-12-15 | 上海前隆信息科技有限公司 | A kind of credit estimation method and system, storage medium and terminal device |
| CN107909274B (en) * | 2017-11-17 | 2023-02-28 | 平安科技（深圳）有限公司 | Enterprise investment risk assessment method and device and storage medium |
| CN107909274A (en) * | 2017-11-17 | 2018-04-13 | 平安科技（深圳）有限公司 | Enterprise investment methods of risk assessment, device and storage medium |
| CN108038103A (en) * | 2017-12-18 | 2018-05-15 | 北京百分点信息科技有限公司 | A kind of method, apparatus segmented to text sequence and electronic equipment |
| CN108038103B (en) * | 2017-12-18 | 2021-08-10 | 沈阳智能大数据科技有限公司 | Method and device for segmenting text sequence and electronic equipment |
| CN109146611B (en) * | 2018-07-16 | 2021-06-15 | 浙江大学 | E-commerce product quality evaluation method and system |
| CN109146611A (en) * | 2018-07-16 | 2019-01-04 | 浙江大学 | A kind of electric business product quality credit index analysis method and system |
| CN110782332A (en) * | 2019-01-18 | 2020-02-11 | 搜信信用产业集团有限公司 | Intelligent credit assessment dynamic tracing method |
| CN112148764A (en) * | 2019-06-28 | 2020-12-29 | 北京百度网讯科技有限公司 | Feature screening method, device, equipment and storage medium |
| CN111125561A (en) * | 2019-11-28 | 2020-05-08 | 泰康保险集团股份有限公司 | Network heat display method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105528465A (en) | Credit status assessment method and device | |
| US11276007B2 (en) | Method and system for composite scoring, classification, and decision making based on machine learning | |
| CN108564286B (en) | Artificial intelligent financial wind-control credit assessment method and system based on big data credit investigation | |
| US20200167869A1 (en) | Real-time predictive analytics engine | |
| CN108711107A (en) | Intelligent financing services recommend method and its system | |
| TW202022769A (en) | Risk identification model training method and device and server | |
| US20120330715A1 (en) | Enhanced systems, processes, and user interfaces for valuation models and price indices associated with a population of data | |
| CN104866969A (en) | Personal credit data processing method and device | |
| CN111401777B (en) | Enterprise risk assessment method, enterprise risk assessment device, terminal equipment and storage medium | |
| CN108596495A (en) | A kind of retail credit business points-scoring system and method | |
| CN105308640A (en) | Methods and systems for automatically generating high quality adverse action notifications | |
| Wagner | Exports, foreign direct investments and productivity: Are services firms different? | |
| CN107437227A (en) | Stock investment analysis apparatus and method | |
| Momeni et al. | Clustering stock market companies via k-means algorithm | |
| US20180330390A1 (en) | Enhanced systems, processes, and user interfaces for targeted marketing associated with a population of assets | |
| CN109345050A (en) | A kind of quantization transaction prediction technique, device and equipment | |
| KR102628559B1 (en) | Method and device for providing real estate mortgage loan automatic screening platform | |
| Hafez et al. | Applying large-scale optimization to evaluate pavement maintenance alternatives for low-volume roads using genetic algorithms | |
| CN107609771A (en) | A kind of supplier's value assessment method | |
| Chen et al. | Bankruptcy prediction using machine learning models with the text-based communicative value of annual reports | |
| CN111210321B (en) | Risk early warning method and system based on contract management | |
| Pan | The use of genetic programming for the construction of a financial management model in an enterprise | |
| CN109063944A (en) | City banking index analysis method and device based on big data analysis technology | |
| CN115293867A (en) | Financial reimbursement user portrait optimization method, device, equipment and storage medium | |
| CN103886473A (en) | Method and system for determining network transaction article suppliers |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160427 |
|
| RJ01 | Rejection of invention patent application after publication |