WO2019154305A1 - 进行学校申请的个人竞争力智能评估系统及方法 - Google Patents

进行学校申请的个人竞争力智能评估系统及方法 Download PDF

Info

Publication number
WO2019154305A1
WO2019154305A1 PCT/CN2019/074358 CN2019074358W WO2019154305A1 WO 2019154305 A1 WO2019154305 A1 WO 2019154305A1 CN 2019074358 W CN2019074358 W CN 2019074358W WO 2019154305 A1 WO2019154305 A1 WO 2019154305A1
Authority
WO
WIPO (PCT)
Prior art keywords
competitiveness
applicant
school
current
data
Prior art date
Application number
PCT/CN2019/074358
Other languages
English (en)
French (fr)
Inventor
莫凌峰
Original Assignee
藕丝科技（深圳）有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 藕丝科技（深圳）有限公司 filed Critical 藕丝科技（深圳）有限公司
Publication of WO2019154305A1 publication Critical patent/WO2019154305A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/20—Education
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/20—Education
    • G06Q50/205—Education administration or guidance

Definitions

  • the present invention relates to the field of information processing technologies, and in particular, to a personal competitiveness intelligent evaluation system and method for applying for a school.
  • the main object of the present invention is to provide an intelligent evaluation system and method for personal competitiveness for applying for a school, aiming at solving the problem that the existing school application consulting tool cannot analyze the individual competitiveness of the school application in combination with the unstructured data of the applicant. .
  • the present invention provides a personal competitiveness intelligent evaluation system for applying for a school, comprising: a user terminal, a connection network, a database, and further includes: a data acquisition subsystem, a competitiveness scoring subsystem, data modeling and analysis. Subsystem, school selection analysis and application subsystem; among them,
  • the data collection subsystem is configured to collect scoring element data of a past applicant, and connect to the user terminal through the connection network, and receive the scoring element data of the current applicant sent by the user terminal, and The scoring element data of the past applicant and the scoring element data of the current applicant are stored in the database;
  • the competitiveness scoring subsystem is configured to determine a total score of the comprehensive competitiveness standard of the current applicant and a competitiveness level according to the scoring element data of the past applicant and the scoring element data of the current applicant;
  • the data modeling and analysis subsystem includes a modeling module and a success analysis module; the modeling module is configured to establish a candidate success model of the pre-selected school according to the scoring element data of a past applicant of the pre-selected school.
  • the success rate analysis module is configured to calculate, according to the scoring element data of the current applicant and the applicant successor model of the pre-selected school, the current applicant's accepted probability of the pre-selected school;
  • the school selection analysis and application subsystem is configured to determine a current applicant's competitiveness improvement strategy corresponding to the pre-selected school; and generate a current school candidate's school selection analysis report.
  • the scoring element items corresponding to the scoring element data include: background information, academic ability, extracurricular activities, standardized examinations, and other preset materials.
  • the competitive scoring subsystem includes a competitive total score module and a competitiveness level module; wherein
  • the competitive total score module is configured to calculate, according to the scoring element data of the past applicant and the scoring element data of the current applicant, the standard scores of each scoring element of the past applicant, and each of the current applicants.
  • the standard score of the scoring element is calculated according to the standard scores of the scoring elements of the past applicants and the standard scores of the current applicants;
  • the competitiveness level module is configured to determine the competitiveness level of the current applicant according to the total score of the comprehensive competitiveness standard of the past applicant and the total score of the comprehensive competitiveness standard of the current applicant; and according to the past The applicant's score factor data determines the level of strength of each school.
  • the school selection analysis and application subsystem comprises a school selection analysis module;
  • the school selection analysis module is configured to calculate an increase rate of the accepted probability of the pre-selected school by changing the score factor data of the current applicant; and determine an optimal liftable plan corresponding to the pre-selected school according to the promotion rate of the accepted probability Sub-factor items and corresponding competitiveness improvement strategies.
  • the school selection analysis module is further configured to determine, according to the applicant successor model of the pre-selected school and the scoring element data of the current applicant, the weakness of the current applicant's application successor model corresponding to the pre-selected school. Divide the element items and determine the corresponding competitiveness improvement strategy.
  • the current applicant's selection analysis report includes at least the current applicant's accepted probability of the pre-selected school, and/or the competitiveness improvement strategy, and/or the comprehensive competitiveness standard total score, and/or the competitiveness level. .
  • the data modeling and analysis subsystem further includes a model optimization module; wherein
  • the model optimization module is configured to determine a competitiveness prediction result according to the application successor model based on preset test data, and obtain a competitiveness result determined by the competitiveness score subsystem based on the preset test data; Comparing the competitiveness prediction result with the competitiveness result, and evaluating the accuracy of the application successful person model according to the obtained comparison result to obtain an accuracy evaluation result; and optimizing the result according to the accuracy evaluation result Apply for a successful person model.
  • the present invention also provides an intelligent evaluation method for personal competitiveness of a school application, including:
  • the step of determining the current applicant's comprehensive competitiveness standard total score and the competitiveness level according to the scoring element data of the past applicant and the scoring element data of the current applicant includes:
  • the step of determining a current applicant's competitiveness promotion strategy corresponding to the pre-selected school specifically includes: the school selection analysis module calculates an increase in the accepted probability of the corresponding pre-selected school by changing the current applicant's scoring element data. Rate; according to the rate of increase of the probability of admission, determine the optimal upgradeable factor items corresponding to the pre-selected schools and the corresponding competitiveness improvement strategies;
  • An intelligent self-competitive evaluation system and method for applying for a school application integrates key unstructured data and structured data highly related to the application for study into the scoring element data, and collects a large amount of data in the past.
  • the scoring element data of the applicant Based on the scoring element data of a large number of past applicants and the quantification of unstructured data in the scoring element data, the current applicant's competitiveness level and the probability of admission to schools of different strength levels are accurately analyzed.
  • the consulting tools applied by the existing schools cannot combine the applicant's unstructured data to make the individual's application for the school's competitiveness analysis insufficient.
  • Intuitive school-choice analysis report which provides applicants with sufficient intuitive, accurate and effective personal competitiveness assessment information for school application, which can be used for decision-making reference, which helps applicants make better school application decisions and increase application success. rate.
  • the embodiment of the invention realizes the one-stop intelligent online school application consultation, and the traditional school application consultation requires a high degree of face-to-face or online communication with the consultant, which saves the complicated process of the applicant and the consultant and the high communication cost, and helps.
  • the applicant has realized the self-service online competitiveness analysis, identified the shortcomings and provided improvement strategies, online application and application result tracking, etc.
  • FIG. 1 is a block diagram of an intelligent evaluation system for personal competitiveness of a school application according to the present invention
  • FIG. 2 is a flow chart for realizing an intelligent evaluation of personal competitiveness of a school application according to the present invention
  • FIG. 3 is a schematic diagram of implementation of a school selection analysis report generated by a personal competitiveness intelligent evaluation system applied for by a school according to the present invention
  • FIG. 4 is a schematic flow chart of a first embodiment of a method for intelligently evaluating a personal competitiveness of a school application according to the present invention
  • FIG. 5 is a schematic flow chart of a second embodiment of a method for intelligently evaluating a personal competitiveness of a school application according to the present invention.
  • the present invention provides a solution for providing a personal competitiveness intelligent evaluation system and method for applying for a school.
  • the application of the personal competitiveness intelligent assessment system and method for applying for a school in the embodiments of the present invention includes, but is not limited to, (1) applying for an intelligent assessment of the personal competitiveness of the local school; and (2) applying for a foreign school. Intelligent assessment of personal competitiveness.
  • the school is a broad concept, such as various elementary schools, junior high schools, high schools, universities, colleges, research institutes, hospitals, and learning and training institutions.
  • the type of school application also includes a transfer application.
  • a personal competitiveness intelligent evaluation system for applying for a school in the present invention includes: a user terminal 100, a connection network 200, a database 300, and a server 400, and the personal competitiveness intelligent evaluation system for applying for a school is further
  • the data acquisition subsystem 500, the competitiveness scoring subsystem 600, the data modeling and analysis subsystem 700, the school selection analysis and application subsystem 800 are included.
  • the user terminal 100 can be a smart phone, a tablet computer, a notebook computer, a desktop computer, or other electronic device. Through the user terminal 100, the user can implement related processing of data, for example, at a specific operation interface of the user terminal, the current applicant inputs. Upload, retrieve, retrieve, and modify specific data, such as scoring element data. The meaning of specific scoring element data is further explained below.
  • the specific data and automatically collected data input and modified by the user are stored in the database 300; in addition, the modification authority to the stored data can be controlled according to different levels and role settings of the user.
  • the connection network 200 can be a packet switched network for communicating communication information from a device to other devices, and can also provide telephone or internet services.
  • Database 300 can be a localized database or a distributed cloud storage database.
  • the server 400 may be a web server having a processor and a memory, and the number may be one or more.
  • the data collection subsystem 500 is configured to collect the scoring element data of the past applicant, and connect to the user terminal 100 through the connection network 200, and receive the scoring element data of the current applicant sent by the user terminal 100, And storing the scoring element data of the past applicant and the scoring element data of the current applicant to the database 300;
  • the scoring element data is a key assessment data of the applicant applying for the school, and generally includes general critical assessment data of the applicants for each type of school.
  • the scoring element data corresponding to the scoring element data includes: background information, academic ability, extracurricular activities, standardized examinations, and other preset materials.
  • the scoring element data involved in the embodiments of the present invention includes structured data and unstructured data. Structured data is data that can be directly quantified, such as the scores of standardized tests; unstructured data is data that cannot be directly quantified, that is, it needs to be artificially transformed to become quantitative data, such as volunteer activities, work experience and other extracurricular activities data. .
  • Each of the scoring element items contains different numeracy elements, which can be customized and increased or decreased according to the specific school type or assessment needs, including but not limited to the following:
  • Background information including at least: the academic reputation of high school, former university, and university faculty;
  • Extracurricular activities including club activities, sports activities, artistic talents, leadership activities, preparatory courses, volunteer experience, work experience;
  • Standardized exams including AP/Honor scores, TOEFL scores, IELTS scores, SAT scores, SAT professional tests (SAT) Subject), ACT score, GRE score, GRE professional test score, GMAT score, LSAT score and other language and professional test scores;
  • Other preset materials including but not limited to: awards, personal progress trends, letters of recommendation, cover letters.
  • the award status is used to reflect the importance of the applicant's award;
  • the personal progress trend is used to reflect the applicant's academic/academic progress and improvement;
  • the recommendation letter is used to reflect the external evaluation of the applicant;
  • the cover letter is used to reflect The applicant's description and self-evaluation of himself.
  • the above-mentioned past applicants include applicants who have applied for the same type of school in the past years. For example, in the competitive intelligence assessment of the current applicant's application for overseas study, the past applicants should select applicants who have applied for overseas study in the past; For current applicants to conduct a competitive intelligence assessment of their national university master's application, past applicants should select applicants who have previously applied for a master's degree in their own university. If a past applicant is successfully admitted to a school, the past applicant is marked as “Applicant Successful”.
  • the scoring element data of the current applicant is the scoring element data uploaded by the current applicant through the user terminal 100.
  • the data collection subsystem 500 includes a data pushing module 510, a data collecting module 520, and a data receiving module 530.
  • the data pushing module 510 is configured to push the predefined scoring element category to the user terminal 100. In this way, the current applicant inputs the relevant scoring element data according to the pushed scoring element category by the user terminal 100, and the user terminal 100 The related scoring element data is uploaded to the data receiving module 530.
  • the data collection module 520 is configured to collect scoring element data of past applicants over the years.
  • the scoring element data of past applicants over the years includes structured scoring element data and unstructured scoring element data of past applicants over the years.
  • the publicly available data of the past applicants also includes school public data.
  • the school public data includes the structured data and unstructured data of the past years of the school, such as the school's geographical location, tuition fees, admission conditions, academic reputation of the department, the number of students admitted in the calendar year, the background data of the applicants, Academic ability data, extracurricular activity data, standardized test data, and other preset material data.
  • the data collection module 520 uses the big data collection technology to collect and analyze the score element data of the applicants over the years.
  • the data receiving module 530 is configured to receive the scoring element data sent by the user terminal 100. Further, the scoring element data of the past applicant and the scoring element data of the current applicant are stored in the database 300; when the scoring element data needs to be called, the scoring element data stored in the database 300 may be called.
  • the competitiveness scoring subsystem 600 is configured to determine a total score of the comprehensive competitiveness standard of the current applicant and a competitiveness level according to the scoring element data of the past applicant and the scoring element data of the current applicant;
  • the competitiveness scoring subsystem 600 includes a competitiveness total score module 610 and a competitiveness level module 620.
  • the competitive total score module 610 is configured to calculate, according to the scoring element data of the past applicant and the scoring element data of the current applicant, the standard scores of the scoring elements of the past applicants, and the current applicants.
  • the standard scores of each scoring element; according to the standard scores of the scoring elements of the past applicants and the scores of the current applicants, the total scores of the comprehensive competitiveness standards of the past applicants and the comprehensive competition of the current applicants are respectively calculated.
  • the total score of the force standard and determining the current applicant's ranking in the preset group according to the total score of the comprehensive competitiveness standard of the current applicant;
  • each of the molecular element items in the scoring element item is hierarchically divided, and a preferred division condition is shown in Table 1.
  • the rankings (including the rankings of senior high school students, university rankings, and academic reputations of colleges and universities) have been published by relevant educational authority rating agencies. If the current applicant has been ranked in the top 60 in the high school, the current applicant's high school ranking item belongs to the fourth level, and other levels are deduced by analogy.
  • GPA Gram Point Average
  • the conversion method is to convert the scores obtained by each subject into grade points, and then perform weighted summation according to the proportion of credits of each subject. The results are divided by the subjects. The sum of credits gives you a grade point average.
  • the four-point GPA system is used in Table 1, including the first level (GPA equals 3.8 or above), the second level (GPA equals 3.7 or above), the third level (GPA equals 3.6 or above), and the fourth level ( GPA is equal to 3.5 or above), fifth level (GPA is equal to 3.3 or above), sixth level (GPA is equal to 3.2 or above), and seventh level (GPA is less than 3.2).
  • Each level data of each sub-element in Table 1 is assigned a corresponding gradation parameter and a weight coefficient, and the scores of the respective numerator elements of the current applicant's score element item Pi(e) are obtained by weighted summation.
  • the calculation formula is:
  • the standard scores of the current applicants are calculated by the conventional evaluation method of the standard scores Pi(es).
  • the conventional method of obtaining is the prior art, and will not be repeated here.
  • Each scoring element score Pi(e) is multiplied by one scoring element weight coefficient to obtain each product respectively; and the respective products are added and summed, and the obtained result is the total competitive total score P(t).
  • the weight factor of the specific scoring element is not limited. In this way, the total competitiveness P(t) of the current applicant (or past applicants) can be obtained.
  • the total score of the comprehensive competitiveness standard P(ts) of the current applicant (or past applicants) can be obtained by calculating the standard score.
  • the current applicant's comprehensive competitiveness standard score P(ts) can objectively reflect the current applicant's level of competitiveness.
  • the current applicant's comprehensive competitiveness standard total score P(ts) the current applicant's ranking in the preset group is determined.
  • the current applicant's ranking among past applicants can be easily obtained.
  • the ranking of applicants in the current year, and then conveniently determine the current level of competitiveness of the applicant relative to the applicant of the year provide reference information for the school application of the year.
  • the competitiveness level module 620 is configured to determine the competitiveness level of the current applicant according to the total score of the comprehensive competitiveness standard of the past applicant and the total score of the comprehensive competitiveness standard of the current applicant. ;
  • the competitiveness level module 620 divides the total scores of the comprehensive competitiveness standard P(ts) of all past applicants into different score ranges; wherein different score ranges correspond to different Competitiveness levels, such as poor, average, strong, strong, and extremely strong.
  • the scores of the total scores of the comprehensive competitiveness standards P(ts) of all past applicants are divided according to a preset ratio. After calculating the current applicant's comprehensive competitiveness standard total score P(ts), the corresponding score range is matched to determine the current applicant's competitiveness level.
  • the competitive level module 620 is further configured to determine a strength level of each school according to the scoring element data of a past applicant;
  • the average value of the standard total score P(ts) is divided into different score ranges; among them, different score ranges correspond to different strength level levels, such as five grades A, B, C, D, and E. . It should be noted that since the score factor data of each applicant's annual application success may change, when determining the strength level of each school, it is necessary to regularly update the scoring element data of the successful applicants of the school over the years, thereby Ensure the accuracy of the school's strength level.
  • the data modeling and analysis subsystem 700 includes a modeling module 710 and a success analysis module 720.
  • the modeling module 710 is configured to establish a pre-selected school according to the scoring element data of a past applicant of a pre-selected school. Applicant's successful applicant model; among them, the pre-selected school is the school selected by the current applicant or the system recommended by the school, that is, the pre-selected schools will conduct intelligent evaluation of the individual competitiveness of the school application, and finally get the current applicant corresponding to each pre-selected school. The results of the individual competitiveness assessment.
  • the modeling module 710 establishes an applicant successor model of the pre-selected school according to the scoring element data of the past applicant and the preset modeling algorithm; wherein the preset modeling algorithm preferably uses a Bayesian classification algorithm. That is, the modeling module 710 classifies and stores the scoring element data of the past applicant based on the collected scoring element data of the past applicant, and extracts each scoring of each past applicant by the data mining algorithm. Feature attributes and metrics for data analysis and modeling.
  • the data mining algorithm includes but is not limited to the following algorithms: a decision tree analysis algorithm, a neural network analysis algorithm (neural Network algorithm), clustering analysis algorithm, association rule analysis algorithm, logistic regression analysis algorithm Algorithm).
  • the modeling module 710 analyzes relevant scoring element data (including scoring element attributes and corresponding indicator data) of all past applicants of a school at a certain strength level level by a decision tree analysis algorithm. This leads to the correlation of all factors on the success of the application and the failure of the application. Then, the clustering analysis algorithm is used to analyze the unique common score element data (including the score element attribute and the corresponding indicator data) respectively of the applicant successful applicant and the application loser. Combine the above two types of analysis results and establish a number of application successful person models, and then sample the randomly selected samples in the database of two types of applicants for success and failure to test the accuracy of the successful applicant model and accurately The highest rate applicant successor model is used as an applicant successor model for predicting the probability of admission to the current applicant.
  • relevant scoring element data including scoring element attributes and corresponding indicator data
  • the success rate analysis module 720 is configured to calculate, according to the scoring element data of the current applicant and the application successor model of the pre-selected school, the current applicant's accepted probability corresponding to the pre-selected school;
  • the success analysis module 720 calculates the score component data of the current applicant based on the applicant successor model, and analyzes and obtains the current applicant's accepted probability corresponding to different valid schools.
  • the school selection analysis and application subsystem 800 is configured to determine a current applicant's competitiveness improvement strategy corresponding to the pre-selected school; and generate a current applicant's school selection analysis report.
  • the current applicant's selection analysis report includes at least the current applicant's accepted probability and competitiveness improvement strategy corresponding to the pre-selected school.
  • the intelligent evaluation system of the present invention can calculate the current applicant's comprehensive competitiveness standard total score P(ts) based on the scoring elements of the current applicant and other applicants (including past applicants and current applicants), The level of competitiveness, and can further obtain the current applicant's ranking under various ranking categories, so that the current applicants have sufficient and comprehensive information on their own competitiveness level, which is more conducive to decision-making. Therefore, the current applicant's school selection analysis report may also include the comprehensive competitiveness standard score P(ts), the competitiveness level, and the related rankings.
  • the school selection analysis report may include the current applicant's own competitiveness level information (including the comprehensive competitiveness standard total score P(ts), the competitiveness level and the related ranking situation), and the current applicant corresponds to several schools.
  • the accepted probabilities and competitiveness enhancement strategies may include, but are not limited to, html, doc, xml, and pdf.
  • the school selection analysis report may further include a visualization data map/table related to the current applicant's competitiveness analysis result.
  • the school selection analysis report may further include: a school recommendation list arranged in reverse order according to the probability of admission, a recommendation list of schools with similar competitiveness levels, and a corresponding related service provider link of the recommendation scheme.
  • the specifically generated school selection analysis report is shown in Figure 3.
  • the competitiveness improvement strategy only generates the corresponding competitiveness improvement strategy based on the applicant's choice of applying for the A school. In the actual operation, the preselection selected by the applicant is selected. Schools can generate corresponding competitiveness improvement strategies.
  • the school selection analysis and application subsystem 800 further includes a push module 820; the push module 820 is configured to send the school selection analysis report of the current applicant to the user terminal 100.
  • the current user receives the generated school selection analysis report through the user terminal 100, thereby knowing the competitiveness result of the user and the selected target school admission probability information.
  • the school selection analysis and application subsystem 800 further includes an application module 830; the application module 830 is configured to implement a school application operation related to the current applicant. For example, a preliminary review of the application materials submitted by the current applicant; and submitting the current applicant's application materials to the official admission application system; and querying the current applicant's application status.
  • the personal competitiveness intelligent assessment system for applying for a school in this embodiment incorporates key unstructured data and structured data highly relevant to the school application into the scoring element data, and collects the scoring elements of the past applicants in large quantities. data. Based on the scoring element data of a large number of past applicants and the quantification of unstructured data in the scoring element data, the current applicant's competitiveness level and the probability of admission to schools of different strength levels are accurately analyzed.
  • the existing learning application consulting tool cannot combine the applicant's unstructured data to make the individual school application competitive analysis insufficient.
  • it also provides accurate and effective competitiveness improvement strategies for current applicants, as well as convenient and intuitive school selection analysis reports and solutions, thus providing applicants with sufficient intuitive, accurate and effective personal competitiveness assessment information for school applications.
  • the embodiment of the invention realizes the one-stop intelligent online school application consultation, and the traditional school application consultation (such as the study abroad consultation) needs a high degree of face-to-face or online communication with the consultant, which saves the complicated process of the communication between the applicant and the consultant and is high.
  • the communication cost helps the applicant to implement the self-service online competitiveness analysis, identify the deficiencies and provide improvement strategies, online application and application result tracking, etc.
  • the school selection analysis and application subsystem 800 includes a school selection analysis module 810;
  • the school selection analysis module 810 is configured to determine the current applicant's competitiveness improvement strategy corresponding to the pre-selected school, and the corresponding implementation manners are various, for example:
  • the school selection analysis module 810 is configured to calculate a promotion rate of the accepted probability of the pre-selected school by changing the scoring element data of the current applicant; and determine a corresponding pre-selection according to the promotion rate of the accepted probability The school's optimality can improve the scoring factor and the corresponding competitiveness improvement strategy.
  • the change of any score factor data is used to improve the admission probability of the pre-selected school; if the pre-selected school's acceptance probability increase rate is more obvious, the score factor corresponding to the changed score element data is Excellent or optimal to improve the scoring element.
  • the current applicant is indicated to be able to effectively improve the corresponding promotion direction of the pre-selected school admission probability, so as to develop a corresponding competitiveness improvement strategy.
  • the Leadership Activity Index increased from the current 2 points to 3 points, the probability of current applicants being admitted to a pre-selected school increased from 60% to 65%; and the number of published papers (first author) increased from 1 to 2. At the time of the article, the probability of being admitted can be increased to 85%.
  • current applicants should focus on increasing the number of published papers (first author) to increase the probability of being accepted.
  • the other scoring element data are analogized in turn, and the promotion rate corresponding to the accepted probability is obtained, thereby finding the promotion scoring element item which is most beneficial to the current applicant to improve the accepted probability of the corresponding pre-selected school.
  • the school selection analysis module 810 is further configured to determine, according to the application successor model of the pre-selected school and the scoring element data of the current applicant, that the current applicant is successfully applied to the pre-selected school.
  • the weak scoring element of the model and the corresponding competitiveness improvement strategy is further configured to determine, according to the application successor model of the pre-selected school and the scoring element data of the current applicant, that the current applicant is successfully applied to the pre-selected school. The weak scoring element of the model and the corresponding competitiveness improvement strategy.
  • the pre-selected school's applicant success model can generally reflect the group average of successful applicants in the pre-selected schools, including the average standard score of each scoring element (the average of the numerator factor scores of the successful applicants in each pre-selected school) .
  • the current applicant's scoring element data is easier to find out the current applicant's weak scoring element item, and the calculation is convenient and accurate.
  • the current applicants are provided with a constructive competitive improvement strategy (including specific optimization suggestions) to improve the data of the weak scoring elements.
  • the current applicant's TOEFL score is 95, which is significantly lower than the average score of the TOEFL score of 105 for a successful applicant in a pre-selected school, the current applicant should focus on improving the TOEFL score; correspondingly, the generated competitiveness improvement strategy includes There are methodological skills and implementation plans to improve TOEFL scores.
  • the potential score item can be found and the strategic improvement suggestions can be provided accordingly. Enable current applicants to increase the application success rate.
  • the current applicant's competitiveness improvement strategy is generated according to the current applicant's comprehensive competitiveness standard total score, the competitiveness level, and the accepted probability. For example, the lower limit of the total score of the comprehensive competitiveness corresponding to the current competitor's competitiveness level and the better level of the first-level competitiveness level, determine the difference between the two points, and find The current applicant's score is lower than the score factor, thereby generating a competitive improvement strategy. For example, if the current applicant's extracurricular activity scoring element score is low, the extracurricular activity scoring element item is marked as a scoring element item that needs improvement.
  • the current applicant obtains the competitiveness improvement strategy through the user terminal, and needs to upgrade and strengthen the various molecular element items corresponding to the extracurricular activity scoring element item, in order to improve the competitiveness level. Further, the achievability analysis is performed on the generated competitiveness improvement strategy. If the analysis result is that the improvement project has difficulty in realizing the improvement project, the improvement project is marked to inform the current applicant.
  • the school selection analysis module 810 is further configured to determine that the current applicant's competitiveness level matches according to the current applicant's comprehensive competitiveness standard total score, the competitiveness level, and the accepted probability.
  • the level of strength of the school according to the determined level of strength of the school, find the school at the level of the strength level, and generate a competitiveness improvement strategy for the current applicant to be admitted to the school of the strength level. And, find schools that are better than the above-mentioned strength level, and the competitiveness improvement strategy that is accepted by the corresponding schools.
  • the personal competitiveness intelligent evaluation system for applying for the school provides the applicant with a multi-angled constructive and enforceable competitiveness improvement strategy, thereby helping the applicant to understand the competitiveness of the applicant, and Targeted improvement and enhancement activities using the provided competitiveness enhancement strategies will help increase the applicant's application success rate.
  • the data modeling and analysis subsystem 700 further includes a model optimization module 730;
  • the model optimization module 730 is configured to determine, according to the preset test data, the competitiveness prediction result according to the application successor model, and obtain the competitiveness result determined by the competitiveness scoring subsystem 600 based on the preset test data. Comparing the competitiveness prediction result with the competitiveness result, and evaluating the accuracy of the application successful person model according to the obtained comparison result to obtain an accuracy evaluation result; and optimizing the result according to the accuracy evaluation result The applicant success model.
  • the selected evaluation data is prediction test data, preferably the score element data of the past applicants stored in the database 300.
  • the scores of the selected random applicants are calculated and analyzed by the applicant success model to obtain the competitiveness prediction results (including but not limited to the comprehensive competitiveness total score and the competitiveness level).
  • the scoring element data of the selected random past applicants is calculated and analyzed by the competitive total score module 610 and the competitiveness level module 620 in the competitive scoring subsystem 600, and the corresponding competitive results are obtained (including But not limited to the overall competitiveness of the total score and competitiveness level).
  • the competitiveness prediction results are compared with the competitiveness results to judge the gap between the two; according to the gap judgment results, the prediction accuracy of the applicant success model is judged.
  • the model of the applicant success model is optimized or a new applicant success model is established.
  • the model can be evaluated and optimized based on other evaluation indicators such as the operational stability of the model.
  • an embodiment of the present invention provides an intelligent assessment method for personal competitiveness of a school application, including the following steps:
  • Step S10 collecting scoring element data of the past applicant, and receiving the scoring element data of the current applicant, and storing the scoring element data of the past applicant and the scoring element data of the current applicant to the database. ;
  • the scoring element data is the key assessment data of the applicant applying for the school, and generally includes the general critical assessment data of the applicants for each type of school.
  • the scoring element data corresponding to the scoring element data includes: background information, academic ability, extracurricular activities, standardized examinations, and other preset materials.
  • the scoring element data involved in the embodiments of the present invention includes structured data and unstructured data. Structured data is data that can be directly quantified, such as the scores of standardized tests; unstructured data is data that cannot be directly quantified, that is, it needs to be artificially transformed to become quantitative data, such as volunteer activities, work experience and other extracurricular activities data. .
  • Each of the scoring element items contains different numeracy elements, which can be customized and increased or decreased according to the specific school type or assessment needs, including but not limited to the following:
  • Background information including at least: the academic reputation of high school, former university, and university faculty;
  • Extracurricular activities including club activities, sports activities, artistic talents, leadership activities, preparatory courses, volunteer experience, work experience;
  • Standardized exams including AP/Honor scores, TOEFL scores, IELTS scores, SAT scores, SAT professional tests (SAT) Subject), ACT score, GRE score, GRE professional test score, GMAT score, LSAT score and other language and professional test scores;
  • Other preset materials including but not limited to: awards, personal progress trends, letters of recommendation, cover letters.
  • the award status is used to reflect the importance of the applicant's award;
  • the personal progress trend is used to reflect the applicant's academic/academic progress and improvement;
  • the recommendation letter is used to reflect the external evaluation of the applicant;
  • the cover letter is used to reflect The applicant's description and self-evaluation of himself.
  • the above-mentioned past applicants include applicants who have applied for the same type of school in the past years. For example, in the competitive intelligence assessment of the current applicant's application for overseas study, the past applicants should select applicants who have applied for overseas study in the past; For current applicants to conduct a competitive intelligence assessment of their national university master's application, past applicants should select applicants who have previously applied for a master's degree in their own university. If a past applicant is successfully admitted to a school, the past applicant is marked as “Applicant Successful”.
  • the scoring element data of the current applicant is the scoring element data uploaded by the current applicant through the user terminal, wherein the user terminal may be a smart phone, a tablet computer, a notebook computer, a desktop computer or other electronic device, through the user terminal,
  • the user can implement related processing of data, for example, at a specific operation interface of the user terminal, the current applicant inputs, uploads, retrieves, retrieves and modifies specific data, such as scoring element data, and the meaning of the specific scoring element data is further explained below.
  • the specific data and automatically collected data entered and modified by the user are stored in the database; in addition, the modification authority to the stored data can be controlled according to different levels and role settings of the user.
  • the scoring element data of past applicants over the years includes structured scoring element data and unstructured scoring element data of past applicants over the years.
  • the publicly available data of the past applicants also includes school public data.
  • the school public data includes the structured data and unstructured data of the past years of the school, such as the school's geographical location, tuition fees, admission conditions, academic reputation of the department, the number of students admitted in the calendar year, the background data of the applicants, Academic ability data, extracurricular activity data, standardized test data, and other preset material data.
  • a large number of data of the scoring elements of the applicants who have applied for the school over the years are collected and analyzed.
  • scoring element data of the past applicant and the scoring element data of the current applicant are stored in the database; when the scoring element data needs to be called, the scoring element data stored in the database may be called.
  • Step S20 determining, according to the scoring element data of the past applicant and the scoring element data of the current applicant, the current applicant's comprehensive competitiveness standard total score and competitiveness level;
  • Step S21 calculating, according to the scoring element data of the past applicant and the scoring element data of the current applicant, the standard scores of each scoring element of the past applicant, and the standard of each scoring element of the current applicant Minute;
  • Step S22 calculating the total score of the comprehensive competitiveness standard of the current applicant according to the standard scores of the scores of the past applicants and the scores of the current applicants; and the comprehensive competitiveness according to the current applicant
  • the standard total score determines the current applicant's ranking in the default group
  • each of the molecular element items in the scoring element item is hierarchically divided, and a preferred division condition is shown in Table 1.
  • the rankings (including the rankings of senior high school students, university rankings, and academic reputations of colleges and universities) have been published by relevant educational authority rating agencies. If the current applicant has been ranked in the top 60 in the high school, the current applicant's high school ranking item belongs to the fourth level, and other levels are deduced by analogy.
  • GPA Gram Point Average
  • the conversion method is to convert the scores obtained by each subject into grade points, and then perform weighted summation according to the proportion of credits of each subject. The results are divided by the subjects. The sum of credits gives you a grade point average.
  • the four-point GPA system is used in Table 1, including the first level (GPA equals 3.8 or above), the second level (GPA equals 3.7 or above), the third level (GPA equals 3.6 or above), and the fourth level ( GPA is equal to 3.5 or above), fifth level (GPA is equal to 3.3 or above), sixth level (GPA is equal to 3.2 or above), and seventh level (GPA is less than 3.2).
  • Each level data of each sub-element in Table 1 is assigned a corresponding gradation parameter and a weight coefficient, and the scores of the respective numerator elements of the current applicant's score element item Pi(e) are obtained by weighted summation.
  • the calculation formula is:
  • the standard scores of the current applicants are calculated by the conventional evaluation method of the standard scores Pi(es).
  • the conventional method of obtaining is the prior art, and will not be repeated here.
  • Each scoring element score Pi(e) is multiplied by one scoring element weight coefficient to obtain each product respectively; and the respective products are added and summed, and the obtained result is the total competitive total score P(t).
  • the weight factor of the specific scoring element is not limited. In this way, the total competitiveness P(t) of the current applicant (or past applicants) can be obtained.
  • the total score of the comprehensive competitiveness standard P(ts) of the current applicant (or past applicants) can be obtained by calculating the standard score.
  • the current applicant's comprehensive competitiveness standard score P(ts) can objectively reflect the current applicant's level of competitiveness.
  • the current applicant's comprehensive competitiveness standard total score P(ts) the current applicant's ranking in the preset group is determined.
  • the current applicant's ranking among past applicants can be easily obtained.
  • the ranking of the applicants in the current year and then conveniently determine the current level of competitiveness of the applicant relative to the applicant of the year, and provide reference information for the application for the current year.
  • Step S23 determining the competitiveness level of the current applicant according to the total score of the comprehensive competitiveness standard of the past applicant and the total score of the comprehensive competitiveness standard of the current applicant;
  • the competitiveness level module 620 divides the total scores of the comprehensive competitiveness standard P(ts) of all past applicants into different score ranges; wherein different score ranges correspond to different Competitiveness levels, such as poor, average, strong, strong, and extremely strong.
  • the scores of the total scores of the comprehensive competitiveness standards P(ts) of all past applicants are divided according to a preset ratio. After calculating the current applicant's comprehensive competitiveness standard total score P(ts), the corresponding score range is matched to determine the current applicant's competitiveness level.
  • step S24 the level of the strength level of each school is determined according to the score element data of the past applicant.
  • the average value of the standard total score P(ts) is divided into different score ranges; among them, different score ranges correspond to different strength level levels, such as five grades A, B, C, D, and E. . It should be noted that since the score factor data of each applicant's annual application success may change, when determining the strength level of each school, it is necessary to regularly update the scoring element data of the successful applicants of the school over the years, thereby Ensure the accuracy of the school's strength level.
  • Step S30 establishing, according to the score element data of the past applicants of the pre-selected school, a model for applying for success in the pre-selected school; wherein, the pre-selected school is a school selected by the current applicant or a system recommended by the system, that is, a plurality of pre-selected schools are selected.
  • the school conducts a self-assessment intelligent assessment of the application for the school, and finally obtains the individual application competitiveness evaluation result of the current applicant corresponding to each pre-selected school.
  • the applicant successor model of each school is established according to the scoring element data of the past applicant and the preset modeling algorithm; wherein the preset modeling algorithm is preferably a Bayesian classification algorithm. That is, based on the scoring element data of the past applicants, the scoring element data of the past applicants are classified and stored, and then each scoring element attribute and index of each past applicant is extracted by the data mining algorithm.
  • the data mining algorithm includes but is not limited to the following algorithms: a decision tree analysis algorithm, a neural network analysis algorithm (neural Network algorithm), clustering analysis algorithm, association rule analysis algorithm, logistic regression analysis algorithm Algorithm).
  • the decision tree analysis algorithm analyzes relevant scoring element data (including scoring element attributes and corresponding indicator data) of all past applicants of a school at a certain level of strength level, thereby obtaining all The correlation between factors on the success of the application and the failure of the application. Then, the clustering analysis algorithm is used to analyze the unique common score element data (including the score element attribute and the corresponding indicator data) respectively of the applicant successful applicant and the application loser. Combine the above two types of analysis results and establish a number of application successful person models, and then sample the randomly selected samples in the database of two types of applicants for success and failure to test the accuracy of the successful applicant model and accurately The highest rate applicant successor model is used as an applicant successor model for predicting the probability of admission to the current applicant.
  • Step S40 calculating, according to the scoring element data of the current applicant and the application successor model of the pre-selected school, calculating the accepted probability of the current applicant corresponding to the pre-selected school;
  • the current applicant's scoring element data is calculated, and the current applicant's acceptance probability corresponding to different pre-selected schools is obtained.
  • step S50 the current applicant's competitiveness improvement strategy corresponding to the pre-selected school is determined; and the current applicant's school selection analysis report is generated.
  • the current applicant's selection analysis report includes at least the current applicant's accepted probability and competitiveness improvement strategy corresponding to the pre-selected school.
  • the intelligent evaluation system of the present invention can calculate the current applicant's comprehensive competitiveness standard total score P(ts) based on the scoring elements of the current applicant and other applicants (including past applicants and current applicants), The level of competitiveness, and can further obtain the current applicant's ranking under various ranking categories, so that the current applicants have sufficient and comprehensive information on their own competitiveness level, which is more conducive to decision-making. Therefore, the current applicant's school selection analysis report may also include the comprehensive competitiveness standard score P(ts), the competitiveness level, and the related rankings.
  • the school selection analysis report may include the current applicant's own competitiveness level information (including the comprehensive competitiveness standard total score P(ts), the competitiveness level and the related ranking situation), and the current applicant corresponds to several schools.
  • the accepted probabilities and competitiveness enhancement strategies may include, but are not limited to, html, doc, xml, and pdf.
  • the school selection analysis report may further include a visualization data map/table related to the current applicant's competitiveness analysis result.
  • the school selection analysis report may further include: a school recommendation list arranged in reverse order according to the probability of admission, a recommendation list of schools with similar competitiveness levels, and a corresponding related service provider link of the recommendation scheme.
  • the specifically generated school selection analysis report is shown in Figure 3.
  • the competitiveness improvement strategy only generates the corresponding competitiveness improvement strategy based on the applicant's choice of applying for the A school. In the actual operation, the preselection selected by the applicant is selected. Schools can generate corresponding competitiveness improvement strategies.
  • the method further includes transmitting the school selection analysis report of the current applicant to the user terminal.
  • the current user receives the generated school selection analysis report through the user terminal, thereby knowing the competitiveness result of the user and the probability information of the selected target school.
  • the implementation of the school application operation related to the current applicant also includes the implementation of the school application operation related to the current applicant. For example, a preliminary review of the application materials submitted by the current applicant; and submitting the current applicant's application materials to the official admission application system; and querying the current applicant's application status.
  • scoring element data key unstructured data and structured data highly correlated with the school application are included in the scoring element data, and the scoring element data of the past applicants are collected in large quantities. Based on the scoring element data of a large number of past applicants and the quantification of unstructured data in the scoring element data, the current applicant's competitiveness level and the probability of admission to schools of different strength levels are accurately analyzed.
  • the existing school application consulting tool cannot combine the applicant's unstructured data to make the analysis of the individual competitiveness of the school application.
  • it also provides accurate and effective competitiveness improvement strategies for current applicants, as well as convenient and intuitive school-based analysis reports and solutions, thus providing current applicants with sufficient intuitive, accurate and effective personal competitiveness assessment information for school applications.
  • the embodiment of the invention realizes the one-stop intelligent online school application consultation, and the traditional school application consultation (such as the study abroad consultation) needs a high degree of face-to-face or online communication with the consultant, which saves the complicated process of the communication between the applicant and the consultant and is high.
  • the communication cost helps the applicant to implement the self-service online competitiveness analysis, identify the deficiencies and provide improvement strategies, online application and application result tracking, etc.
  • the step of determining a current applicant's competitiveness promotion strategy corresponding to the pre-selected school includes:
  • Step S51 calculating the promotion rate of the accepted probability of the corresponding pre-selected school by changing the scoring element data of the current applicant; determining the optimal upgradeable scoring element item corresponding to the pre-selected school according to the promotion rate of the accepted probability and corresponding Competitive improvement strategy.
  • the change of any score factor data is used to improve the admission probability of the pre-selected school; if the pre-selected school's acceptance probability increase rate is more obvious, the score factor corresponding to the changed score element data is Excellent or optimal to improve the scoring element.
  • the current applicant is indicated to be able to effectively improve the corresponding promotion direction of the pre-selected school admission probability, so as to develop a corresponding competitiveness improvement strategy.
  • the Leadership Activity Index increased from the current 2 points to 3 points, the probability of current applicants being admitted to a pre-selected school increased from 60% to 65%; and the number of published papers (first author) increased from 1 to 2. At the time of the article, the probability of being admitted can be increased to 85%.
  • current applicants should focus on increasing the number of published papers (first author) to increase the probability of being accepted.
  • the other scoring element data are analogized in turn, and the promotion rate corresponding to the accepted probability is obtained, thereby finding the promotion scoring element item which is most beneficial to the current applicant to improve the accepted probability of the corresponding pre-selected school.
  • step of determining a current applicant's competitiveness promotion strategy corresponding to the pre-selected school further includes:
  • Step S52 determining, according to the applicant successor model of the pre-selected school and the scoring element data of the current applicant, the weak scoring element item of the current applicant's applicant success model corresponding to the pre-selected school, and determining the corresponding competitiveness. Improve the strategy.
  • the pre-selected school's applicant success model can generally reflect the group average of successful applicants in the pre-selected schools, including the average standard score of each scoring element (the average of the numerator factor scores of the successful applicants in each pre-selected school) .
  • the current applicant's scoring element data is easier to find out the current applicant's weak scoring element item, and the calculation is convenient and accurate.
  • the current applicants are provided with a constructive competitive improvement strategy (including specific optimization suggestions) to improve the data of the weak scoring elements.
  • the current applicant's TOEFL score is 95, which is significantly lower than the average score of the TOEFL score of 105 for a successful applicant in a pre-selected school, the current applicant should focus on improving the TOEFL score; correspondingly, the generated competitiveness improvement strategy includes There are methodological skills and implementation plans to improve TOEFL scores.
  • the potential score item can be found and the strategic improvement suggestions can be provided accordingly. Enable current applicants to increase the application success rate.
  • the current applicant's competitiveness improvement strategy is generated according to the current applicant's comprehensive competitiveness standard total score, the competitiveness level, and the accepted probability. For example, the lower limit of the total score of the comprehensive competitiveness corresponding to the current competitor's competitiveness level and the better level of the first-level competitiveness level, determine the difference between the two points, and find The current applicant's score is lower than the score factor, thereby generating a competitive improvement strategy for the actual competitiveness of the current applicant.
  • the competitiveness improvement strategy includes: if the current applicant's extracurricular activity scoring element score is low, the extracurricular activity scoring element item is marked as a scoring element item that needs improvement.
  • the current applicant obtains the competitiveness improvement strategy through the user terminal, and should upgrade and strengthen the various molecular element items corresponding to the extracurricular activity scoring element items, in order to improve the competitiveness level. Further, the achievability analysis is performed on the generated competitiveness improvement strategy. If the analysis result is that the improvement project has difficulty in realizing the improvement project, the improvement project is marked to inform the current applicant.
  • the applicant is provided with a multi-angled constructive and enforceable competitiveness improvement strategy, thereby helping the applicant to understand the competitiveness of the applicant and making use of the provided competitiveness improvement strategy.
  • Sexual improvement and promotion activities will help increase the applicant's application success rate.
  • the method further includes:
  • Step S60 Determine, according to the preset test data, a competitiveness prediction result according to the application successor model, and obtain a competitiveness result determined by the competitiveness scoring subsystem based on the preset test data;
  • Step S61 comparing the competitiveness prediction result with the competitiveness result, and evaluating the accuracy of the application successful person model according to the obtained comparison result to obtain an accuracy evaluation result;
  • Step S62 optimizing the application successor model according to the accuracy assessment result.
  • the selected evaluation data is prediction test data, preferably the scoring element data of past applicants stored in the database.
  • the scores of the selected random applicants are calculated and analyzed by the applicant success model to obtain the competitiveness prediction results (including but not limited to the comprehensive competitiveness total score and the competitiveness level).
  • the selected scoring element data of a plurality of random past applicants are calculated and analyzed to obtain corresponding competitiveness results (including but not limited to the comprehensive competitiveness total score and the competitiveness level).
  • the competitiveness prediction results are compared with the competitiveness results to judge the gap between the two; according to the gap judgment results, the prediction accuracy of the applicant success model is judged.
  • the model of the applicant success model is optimized or a new applicant success model is established.
  • the model can be evaluated and optimized based on other evaluation indicators such as the operational stability of the model.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Educational Administration (AREA)
  • Economics (AREA)
  • Tourism & Hospitality (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Theoretical Computer Science (AREA)
  • Development Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Physics & Mathematics (AREA)
  • Educational Technology (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种进行学校申请的个人竞争力智能评估系统及方法，该系统包括：数据采集子系统（500）用于采集过往申请者的计分要素数据及接收当前申请者的计分要素数据；竞争力评分子系统（600）用于确定当前申请者的综合竞争力标准总分及竞争力等级；数据建模与分析子系统（700）用于建立各个学校的申请成功者模型及计算当前申请者对应各个学校的被录取概率；择校分析与申请子系统（800）用于根据当前申请者的综合竞争力标准总分、竞争力等级及被录取概率，生成当前申请者的择校分析报告。解决了现有学校申请咨询工具无法结合申请者的非结构化数据进行申请入学的个人竞争力分析的问题。

Description

进行学校申请的个人竞争力智能评估系统及方法
技术领域
本发明涉及信息处理技术领域，尤其涉及一种进行学校申请的个人竞争力智能评估系统及方法。
背景技术
对于很多人而言，进入学校（如各类大学或者学院）学习或者深造会对人生的发展产生重要的作用。一般而言，进入的学校主要有国内学校和国外学校两大类。而随着国人的生活水平的提高，中国人除了进入国内学校学习以外，赴国外进行学习或者深造已经形成一种新风尚。同时，由于中国的对外开放水平不断提高，与国外的交流联系日益紧密，外国人来中国进行留学或者深造同样成为一种新趋势。
但是，对于很多想进入国内或国外学校学习或者深造的人士，由于自身对信息收集、处理和分析的能力及相关资源有限，往往难以对自身进行学校申请的竞争力进行客观且全面的评估，也无法识别自身在进行学校申请时存在的薄弱环节，更遑论对薄弱环节进行改善和提升以增强个人进行学校申请的竞争力。
以海外留学为例，很多想留学的人士此前并没有接触过出国留学的相关事宜，对出国留学需要准备的事宜毫无头绪。由于对留学的准备不充分以及对国外学校情况的不了解，自主申请留学的申请者出于各种原因往往同时申请多所学校以提高自己被录取的成功率；而现实环境中，其他大部分的非自主申请留学申请者会选择更熟悉留学申请情况的留学服务机构的咨询服务，以期提高留学申请成功率。
对于自主申请留学的申请者而言，由于自身外语能力和对国外情况知之甚少，加上每个学校在审核申请时使用的审核指标又不尽相同，此类申请者经常无法申请到自己满意的学校或者无法申请到奖学金。而对于选择留学服务机构协助申请的申请者，各个留学服务机构的收费标准及申请者竞争力评估标准均不统一，导致申请者往往面临复杂的选择困境和极其高昂的服务获取成本。传统的留学服务机构长期以来通过观察、记录和分析过往的申请结果，帮助申请者分析申请材料以期优化申请结果，这需要相关的从业人员具备非常扎实的专业知识和实践经验，同时需要消耗巨大的人力和物力。
虽然目前市场上传统的留学咨询工具虽然可帮助申请者核查留学考试分数，如GPA（Grade Point Average，平均学分绩点）、TOEFL（The Test of English as a Foreign Language，托福考试）、GRE（Graduate Record Examination，美国研究生入学考试）成绩等，并为申请者选择合适的学校。然而，这些基于结构化数据的咨询工具不能对个人工作背景、课外活动背景、学术背景或者其它非结构化数据进行分析并给出改进意见，用以指导申请者在某一留学申请指标考核领域进行提升。
上述内容仅用于辅助理解本发明的技术方案，并不代表承认上述内容是现有技术。
发明内容
本发明的主要目的在于提供一种进行学校申请的个人竞争力智能评估系统及方法，旨在解决现有学校申请咨询工具无法结合申请者的非结构化数据进行学校申请的个人竞争力分析的问题。
为实现上述目的，本发明提供一种进行学校申请的个人竞争力智能评估系统，包括：用户终端、连接网络、数据库，还包括：数据采集子系统、竞争力评分子系统、数据建模与分析子系统、择校分析与申请子系统；其中，
所述数据采集子系统，用于采集过往申请者的计分要素数据，以及通过所述连接网络与所述用户终端连接，接收所述用户终端发送的当前申请者的计分要素数据，并将过往申请者的所述计分要素数据及当前申请者的所述计分要素数据存储至所述数据库；
所述竞争力评分子系统，用于根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，确定当前申请者的综合竞争力标准总分及竞争力等级；
所述数据建模与分析子系统包括建模模块、成功率分析模块；所述建模模块，用于根据预选学校的过往申请者的所述计分要素数据，建立预选学校的申请成功者模型；所述成功率分析模块，用于根据当前申请者的所述计分要素数据及预选学校的所述申请成功者模型，计算当前申请者对应预选学校的被录取概率；
所述择校分析与申请子系统，用于确定当前申请者对应预选学校的竞争力提升策略；以及生成当前申请者的择校分析报告。
优选地，所述计分要素数据对应的计分要素项包括：背景资料、学术能力、课外活动、标准化考试及其它预设材料。
优选地，所述竞争力评分子系统包括竞争力总分模块、竞争力等级模块；其中，
所述竞争力总分模块，用于根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，分别计算过往申请者各计分要素标准分、当前申请者的各计分要素标准分；根据过往申请者各计分要素标准分、当前申请者的各计分要素标准分，计算当前申请者的所述综合竞争力标准总分；
所述竞争力等级模块，用于根据过往申请者的所述综合竞争力标准总分、当前申请者的所述综合竞争力标准总分，确定当前申请者的所述竞争力等级；以及根据过往申请者的所述计分要素数据，确定各个学校的实力水平等级。优选地，所述择校分析与申请子系统包括择校分析模块；
所述择校分析模块，用于通过改变当前申请者的计分要素数据，计算对应预选学校的被录取概率的提升率；根据被录取概率的提升率，确定对应预选学校的最优可提升计分要素项及对应的竞争力提升策略。
优选地，所述择校分析模块，还用于根据预选学校的所述申请成功者模型及当前申请者的所述计分要素数据，确定当前申请者对应预选学校的申请成功者模型的薄弱计分要素项，以及确定对应的竞争力提升策略。
优选地，当前申请者的所述择校分析报告至少包括当前申请者对应预选学校的被录取概率、及/或竞争力提升策略、及/或综合竞争力标准总分、及/或竞争力等级。
优选地，所述数据建模与分析子系统还包括模型优化模块；其中，
所述模型优化模块，用于基于预设测试数据，根据所述申请成功者模型确定竞争力预测结果，并获取所述竞争力评分子系统基于所述预设测试数据确定的竞争力结果；将所述竞争力预测结果与所述竞争力结果进行比较，并根据得到的比较结果评估所述申请成功者模型的准确度，以获得准确度评估结果；根据所述准确度评估结果，优化所述申请成功者模型。
此外，为实现上述目的，本发明还提供一种进行学校申请的个人竞争力智能评估方法，包括：
采集过往申请者的计分要素数据，以及接收当前申请者的计分要素数据，并将过往申请者的所述计分要素数据及当前申请者的所述计分要素数据存储至数据库；
根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，确定当前申请者的综合竞争力标准总分及竞争力等级；
根据预选学校的过往申请者的所述计分要素数据，建立预选学校的申请成功者模型；
根据当前申请者的所述计分要素数据及预选学校的所述申请成功者模型，计算当前申请者对应预选学校的被录取概率；
确定当前申请者对应预选学校的竞争力提升策略；以及生成当前申请者的择校分析报告。
优选地，所述根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，确定当前申请者的综合竞争力标准总分及竞争力等级的步骤包括：
根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，分别计算过往申请者各计分要素标准分、当前申请者的各计分要素标准分；
根据过往申请者各计分要素标准分、当前申请者的各计分要素标准分，计算当前申请者的所述综合竞争力标准总分；
根据过往申请者的所述综合竞争力标准总分、当前申请者的所述综合竞争力标准总分，确定当前申请者的所述竞争力等级；
以及根据过往申请者的所述计分要素数据，确定各个学校的实力水平等级。
优选地，所述确定当前申请者对应预选学校的竞争力提升策略的步骤，具体包括：所述择校分析模块通过改变当前申请者的计分要素数据，计算对应预选学校的被录取概率的提升率；根据被录取概率的提升率，确定对应预选学校的最优可提升计分要素项及对应的竞争力提升策略；
或者，根据预选学校的所述申请成功者模型及当前申请者的所述计分要素数据，确定当前申请者对应预选学校的申请成功者模型的薄弱计分要素项，以及确定对应的竞争力提升策略。
本发明实施例提出的一种进行学校申请的个人竞争力智能评估系统及方法，将与留学申请高度相关的关键性非结构化数据及结构化数据均纳入至计分要素数据，并大量采集过往申请者的计分要素数据。基于大量过往申请者的计分要素数据，以及计分要素数据中的非结构化数据的量化，准确地分析出当前申请者的竞争力水平及对应不同实力水平等级学校的被录取概率，弥补了现有学校申请的咨询工具无法结合申请者的非结构化数据进行个人申请学校竞争力分析的不足。同时，也为当前申请者提供了与预选学校对应的准确有效的竞争力提升策略、竞争力水平信息（包括综合竞争力标准总分、在预设群体中的排名情况、竞争力等级）以及方便直观的择校分析报告，从而为申请者提供了学校申请的充分直观、准确有效的个人竞争力评估信息，用以进行决策参考，有助于申请者更好地作出学校申请决策及增加申请成功率。本发明实施例实现了一站式智能在线学校申请咨询，与传统学校申请咨询需要高度的面对面或在线与咨询师交流不同，节省了申请者与咨询师交流的繁琐流程和高昂的沟通成本，帮助申请者实现了自助式的线上竞争力分析、找出不足并提供改进策略、在线申请及申请结果追踪等整套线上学校申请准备全过程。
附图说明
图1为本发明进行学校申请的个人竞争力智能评估系统框图；
图2为本发明进行学校申请的个人竞争力智能评估的实现流程图；
图3为本发明进行学校申请的个人竞争力智能评估系统生成的择校分析报告的实现示意图；
图4为本发明进行学校申请的个人竞争力智能评估方法第一实施例的流程示意图；
图5为本发明进行学校申请的个人竞争力智能评估方法第二实施例的流程示意图。
本发明目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
具体实施方式
应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
由于现有学校申请咨询工具无法结合申请者的非结构化数据进行个人申请学校竞争力分析，本发明提供一种解决方案，即提供一种进行学校申请的个人竞争力智能评估系统及方法。
需要说明的是，本发明实施例一种进行学校申请的个人竞争力智能评估系统及方法适用情况包括但不限于：（1）申请本国学校的个人竞争力智能评估；（2）申请国外学校的个人竞争力智能评估。其中，所述学校为一宽泛性概念，例如各类小学、初中、高中、大学、学院、研究所/院、学习培训机构。此外，进行学校申请的类型还包括转学申请。
如图1所示，本发明的一种进行学校申请的个人竞争力智能评估系统包括：用户终端100、连接网络200、数据库300、服务器400，所述进行学校申请的个人竞争力智能评估系统还包括：数据采集子系统500、竞争力评分子系统600、数据建模与分析子系统700、择校分析与申请子系统800。
下面，结合图2所示的进行学校申请的个人竞争力智能评估的实现流程图进行详细说明。
其中，用户终端100可以是智能手机、平板电脑、笔记本电脑、台式电脑或者其它电子装置，通过用户终端100，用户可以实现数据的相关处理，例如在用户终端的特定操作界面，当前申请者进行输入、上传、获取、检索并修改特定数据，例如计分要素数据，具体的计分要素数据含义在下文进一步说明。用户输入和修改的特定数据和自动采集的数据被存储在数据库300中；此外，还可以根据用户的不同级别和角色设置来控制对存储数据的修改权限。
连接网络200可以是分组交换网，用于将通信信息从某一装置传递到其它装置，还可以提供电话或互联网服务。
数据库300可以是本地式数据库或分布式云存储数据库。
服务器400可以是拥有处理器和存储器的网络服务器，数量可以是一个或者多个。
数据采集子系统500，用于采集过往申请者的计分要素数据，以及通过所述连接网络200与所述用户终端100连接，接收所述用户终端100发送的当前申请者的计分要素数据，并将过往申请者的所述计分要素数据及当前申请者的所述计分要素数据存储至所述数据库300；
其中，所述计分要素数据为申请者进行学校申请的关键性考核数据，一般包括各类学校针对申请者的通用关键性考核数据。在一些实施例中，所述计分要素数据对应的计分要素项包括：背景资料、学术能力、课外活动、标准化考试及其它预设材料。需要注意的是，本发明各实施例中涉及的计分要素数据包括结构化数据及非结构化数据。结构化数据为可以直接量化的数据，如标准化考试的成绩分数；非结构化数据为无法直接量化的数据，即需要进行人为转化才可以成为量化数据，例如志愿者经历、工作经历等课外活动数据。
每项计分要素项包含不同的计分子要素项，可根据具体申请的学校类型或者评估需要进行自定义设置和增减，具体包括但不限于如下内容：
背景资料，至少包括：曾就读高中、曾就读大学、曾就读大学的院系学术声誉排名；
学术能力，包括但不限于高中GPA、大学GPA、大学专业GPA、年级/班级排名、荣誉课程(AP/Honor)、A-G科目（A为历史、社会科学，B为英语，C为数学，D为实验室科学，E为除英语之外的语言，F为视觉与表演艺术，G为大学预科选修课）、科研经历、发表论文情况；
课外活动，包括社团活动、体育活动、艺术才能、领导活动、预科情况、志愿者经历、工作经历；
标准化考试，包括AP/Honor成绩、TOEFL成绩、IELTS成绩、SAT成绩、SAT专业测试(SAT Subject)成绩、ACT成绩、GRE成绩、GRE专业测试成绩、GMAT成绩、LSAT成绩及其他语言和专业性测试成绩；
其他预设材料，包括但不限于：获奖情况、个人进步趋势、推荐信、自荐信。其中，获奖情况用于反映申请者获得奖项的重要程度；个人进步趋势用于反映申请者的学业/学术的进步和提升情况；推荐信用于反映外界对于申请者的个人评价；自荐信用于反映申请者对于自身的描述和自我评价。举例来说，若申请者在本科学习期间分别获得省级、国家级、国际级共三个学术奖项，且上述三个奖项的学术重要性依次递增，则在评价申请者的获奖情况时，给予不同学术重要性的学术奖项对应的分数，并依照奖项重要性设置相应权重，获得申请者在获奖情况要素的总分，从而对申请者的获奖情况进行量化计算及分析。
上述过往申请者包括历年曾经进行同一类学校申请的申请者，例如在对当前申请者进行海外留学申请的竞争力智能评估时，过往申请者应选择过往曾经进行过海外留学申请的申请者；在对当前申请者进行本国大学硕士申请的竞争力智能评估时，过往申请者应选择过往曾经进行过本国大学硕士申请的申请者。若某一过往申请者成功被某一学校录取，则将该过往申请者标记为“申请成功者”。当前申请者的计分要素数据为当前申请者通过用户终端100上传的计分要素数据。
具体地，数据采集子系统500，包括：数据推送模块510、数据收集模块520、数据接收模块530。其中，数据推送模块510用于将预定义的计分要素类别推送至用户终端100；这样，当前申请者通过用户终端100根据推送的计分要素类别输入相关的计分要素数据，用户终端100将相关的计分要素数据上传至数据接收模块530。
数据采集模块520，用于采集历年过往申请者的计分要素数据。历年过往申请者的计分要素数据包括历年过往申请者的结构化计分要素数据和非结构化计分要素数据。过往申请者的计分要素数据中还包括学校公开数据。其中，学校公开数据包括对应学校公布的历年录取的结构化数据和非结构化数据，例如学校地理位置、学费情况、录取条件、院系学术声誉排名、历年录取人数、录取者的背景资料数据、学术能力数据、课外活动数据、标准化考试数据以及其他预设材料数据。具体地，数据采集模块520运用大数据采集技术，大量采集并分析历年申请者的计分要素数据。
数据接收模块530，用于接收用户终端100发送的计分要素数据。此外，历年过往申请者的计分要素数据和当前申请者的计分要素数据存储于数据库300；当需要调用计分要素数据时，调用数据库300中存储的上述计分要素数据即可。
所述竞争力评分子系统600，用于根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，确定当前申请者的综合竞争力标准总分及竞争力等级；
具体地，竞争力评分子系统600包括竞争力总分模块610、竞争力等级模块620。
所述竞争力总分模块610，用于根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，分别计算过往申请者各计分要素标准分、当前申请者的各计分要素标准分；根据过往申请者各计分要素标准分、当前申请者的各计分要素标准分，分别计算过往申请者的综合竞争力标准总分、当前申请者的所述综合竞争力标准总分；以及根据当前申请者的所述综合竞争力标准总分，确定当前申请者在预设群体中的排名情况；
下面，结合一优选实施例（以个人申请国外大学为例）进行说明。
（1）根据预设划分规则，将计分要素项中的各计分子要素项进行层次划分，一种优选的划分情况如表1所示。
表1
计分要素 计分 子要素 数据划分层次以及划分规则
第一层 第二层 第三层 第四层 第五层 第六层 第七层
背景资料 曾就读高中 排名 前10名 前30名 前50名 前70名 前100名 前150名 前150名以外
曾就读大学 排名 前10名 前30名 前50名 前70名 前100名 前150名 前150名以外
曾就读大学的院系学术声誉排名 前10名 前30名 前50名 前70名 前100名 前150名 前150名以外
学术能力 高中GPA 3.8 3.7 3.6 3.5 3.3 3.2 低于3.2
大学GPA 3.8 3.7 3.6 3.5 3.2 3 低于3
专业GPA 3.9 3.8 3.7 3.6 3.5 3.4 低于3.4
年级/班级排名 1% 3% 5% 7% 10% 12% 低于12%
荣誉课程(AP/Honor) 20 18 16 12 8 6 低于6
A-G 科目 56 54 52 50 48 46 低于46
科研经历 3 2 1 / / / /
发表论文数量（第一作者） 2 1 / / / / /
发表论文数量（第二作者） 3 2 1 / / / /
课外活动 社团活动 6 5 4 3 2 1 0
体育活动 6 5 4 3 2 1 0
艺术才能 3 2 1 / / / /
领导活动 3 2 1 / / / /
预科 3 2 1 / / / /
志愿者（小时） 500 400 300 200 100 50 低于50
工作经历 6 5 4 3 2 1 0
标准化考试 TOEFL成绩 118 115 110 102 94 80 低于80
IELTS成绩 9 8.5 8 7.5 7 6.5 低于6.5
SAT成绩 1500 1450 1400 1350 1250 1150 低于1150
SAT Subject成绩平均分 765 750 720 700 680 650 低于650
ACT成绩 33 32 30 29 26 23 低于23
GRE成绩 325 320 310 300 290 280 低于280
GRE Subject成绩 980 970 960 950 920 900 低于900
GMAT成绩 730 700 680 650 620 600 低于600
LSAT成绩 175 170 165 160 155 150 低于150
其他预设材料 获奖情况 / / / / / / /
个人进步趋势 / / / / / / /
举例来说，在教育背景项中，各项排名（包括曾就读高中排名、曾就读大学排名、曾就读大学的院系学术声誉排名）为相关教育权威评级机构公布的综合排名。若当前申请者的曾就读高中排名为前60名，则当前申请者的曾就读高中排名项属于第四层，其它的层次划分以此类推。又例如，GPA(Grade Point Average)即平均绩点，是用于评估学生成绩的一个指标，换算方法为把各个学科所得到的成绩换算为绩点，再按照各学科的学分比例进行加权求和，所得结果除以各学科学分总和即可得到平均绩点。表1中采用的是四分制GPA系统，包括第一层次（GPA等于3.8或以上）、第二层次（GPA等于3.7或以上）、第三层次（GPA等于3.6或以上）、第四层次（GPA等于3.5或以上）、第五层次（GPA等于3.3或以上）、第六层次（GPA等于3.2或以上）、第七层次（GPA小于3.2）。
将表1中每个子要素的每一层次数据赋予对应的层次参数和权重系数，通过加权求和的方式求取当前申请者某一计分要素项的各计分子要素分数Pi(e)，其计算公式为：
Pi(e)=∑ (每个计分子要素最大值*权重系数*层次参数)
然后，基于当前申请者某一计分要素项的各计分子要素分数Pi(e)，通过标准分数的常规求取方式计算当前申请者的各计分要素标准分Pi(es)。常规求取方式为现有技术，此处不再一一赘述。
然后，根据各计分要素分数Pi(e)，计算得到综合竞争力总分P(t)。具体计算方式如下：
将各计分要素分数Pi(e)分别乘以一个计分要素权重系数，分别得到各个乘积；将各个乘积相加求和，所得结果即为综合竞争力总分P(t)。具体的计分要素权重系数不作限制。这样，就可以得到当前申请者（或者过往申请者）的综合竞争力总分P(t)。
这样，就可以通过计算标准分数的方式得到当前申请者（或者过往申请者）的综合竞争力标准总分P(ts)。当前申请者的综合竞争力标准总分P(ts)可以客观地反映出当前申请者的竞争力水平。
然后，根据当前申请者的综合竞争力标准总分P(ts)，确定当前申请者在预设群体中的排名情况。
具体地，根据当前申请者的所述综合竞争力标准总分及纳入计算的申请者（过往申请者加上当前申请者）总数，可以很容易地获得当前申请者在过往申请者中的排名情况，从而精确地刻划出当前申请者相对于过往申请者的竞争力水平。此外，还可以通过获取与当前申请者同年进行学校申请的多数或者全部其它申请者的计分要素数据，并计算出当前申请者在当年申请的综合竞争力标准总分，从而获得当前申请者在当年申请者当中的排名情况，进而方便地确定出当前申请者相对于当年申请者的竞争力水平，提供当年学校申请的参考信息。
进一步地，所述竞争力等级模块620，用于根据过往申请者的所述综合竞争力标准总分、当前申请者的所述综合竞争力标准总分，确定当前申请者的所述竞争力等级；
在一些具体实施中，竞争力等级模块620对所有过往申请者的综合竞争力标准总分P(ts)进行分值范围划分，分成不同的分值区间；其中，不同的分值区间对应不同的竞争力等级，例如较差、一般、较强、很强、极强五个等级。优选地，按照预设比例对所有过往申请者的综合竞争力标准总分P(ts)进行分值范围划分。在计算出当前申请者的综合竞争力标准总分P(ts)后，匹配出对应的分值范围，从而确定当前申请者的竞争力等级。
进一步地，所述竞争力等级模块620，还用于根据过往申请者的所述计分要素数据，确定各个学校的实力水平等级；
即在确定某一学校的实力水平等级时，从所有过往申请者中筛选出该学校历年的申请成功者，并按照上述方法依次求取该学校历年的申请成功者的各计分子要素分数Pi(e)、综合竞争力标准总分P(ts)及竞争力等级。在确定各个学校的实力水平等级时，优选地，先计算各个学校历年的申请成功者的综合竞争力标准总分P(ts)的平均值，然后对各个学校历年的申请成功者的综合竞争力标准总分P(ts)的平均值进行分值范围划分，分成不同的分值区间；其中，不同的分值区间对应不同的实力水平等级，例如A、B、C、D、E五个等级。需要说明的是，由于每个学校每年的申请成功者的计分要素数据可能会变化，因此在确定各所学校的实力水平等级时，需定期更新学校历年的申请成功者的计分要素数据，从而保证学校实力水平等级的准确性。
此外，除了上述确定各个学校的实力水平等级的方法以外，还可以引用权威机构发布的学校实力水平等级评定结果确定各个学校的实力水平等级。
所述数据建模与分析子系统700包括建模模块710、成功率分析模块720；所述建模模块710，用于根据预选学校的过往申请者的所述计分要素数据，建立预选学校的申请成功者模型；其中，预选学校为当前申请者选定的学校或者系统推荐的学校，即通过预选若干个学校进行学校申请的个人竞争力智能评估，最终得到与各预选学校对应的当前申请者的个人竞争力评估结果。
在一些具体实施中，建模模块710根据过往申请者的计分要素数据及预设建模算法建立预选学校的申请成功者模型；其中，预设建模算法优选贝叶斯分类算法。即建模模块710以采集的过往申请者者的计分要素数据为基础，将过往申请者的计分要素数据进行分类存储，再通过数据挖掘算法提取出每个过往申请者的每个计分要素属性和指标，以便进行数据分析和建模。所述数据挖掘算法包括但不限于以下算法：决策树分析算法、神经网络分析算法（neural network algorithm）、聚类分析算法、关联规则分析算法、逻辑斯谛回归分析算法（Logistic regression algorithm）。
在一些具体实施中，建模模块710通过决策树分析算法将处于某一实力水平等级的学校的所有过往申请者的相关计分要素数据（包括计分要素属性及对应的指标数据）进行分析，从而得出所有因素对申请成功和申请失败的影响相关性。然后，通过聚类分析算法分析出申请成功者和申请失败者分别所具有的特有共同计分要素数据（包括计分要素属性及对应的指标数据）。将以上两类分析结果进行组合排序并建立若干个申请成功者模型，再通过对申请成功和失败者两类人的数据库中随机抽取样本进行取样，测试申请成功者模型的准确率，并将准确率最高的申请成功者模型作为用于对当前申请者进行被录取概率预测的申请成功者模型。
进一步地，所述成功率分析模块720，用于根据当前申请者的所述计分要素数据及预选学校的所述申请成功者模型，计算当前申请者对应预选学校的被录取概率；
在一些具体实施中，成功率分析模块720基于申请成功者模型，对当前申请者的计分要素数据进行运算，分析得出当前申请者对应不同有效学校的被录取概率。
进一步地，所述择校分析与申请子系统800，用于确定当前申请者对应预选学校的竞争力提升策略；以及生成当前申请者的择校分析报告。
确定当前申请者对应预选学校的竞争力提升策略，目的在于：向申请者提供多角度的具有建设性及可执行性的竞争力提升策略，从而有助于申请者了解自身的竞争力情况，并利用提供的竞争力提升策略作出针对性的改进及提升活动，有助于增加申请者的申请成功率。
在一些具体实施中，当前申请者的所述择校分析报告至少包括当前申请者对应预选学校的被录取概率和竞争力提升策略。
需要注意的是，由于很多学校设有申请学生录取限额，以及很多学校对于特定国家的留学生也设有录取限额。因此，只提供当前申请者对应预选学校的被录取概率，并不能给当前申请者提供充分全面的参考信息。本发明与现有的传统人工咨询方式或者在线咨询系统进行咨询的方式的不同之处在于，本发明的智能评估系统不仅向当前申请者提供直观的个人综合排名以及在其所处竞争力等级的排名，还能通过刷选的方式向当前申请者提供在本国当年申请者中的排名。也即，本发明的智能评估系统可以基于当前申请者及其他申请者（包括过往申请者和当年申请者）的计分要素，计算出当前申请者的综合竞争力标准总分P(ts)、竞争力等级，并可以进一步地得到当前申请者在各类排名类别下的排名情况，从而使得当前申请者掌握充分且全面的自身竞争力水平信息，从而更有利于决策参考。因此，当前申请者的择校分析报告中还可以包括综合竞争力标准总分P(ts)、竞争力等级及相关的排名情况。
即所述择校分析报告中可包含当前申请者的自身竞争力水平信息（包括综合竞争力标准总分P(ts)、竞争力等级及相关的排名情况）、当前申请者对应若干个学校的被录取概率、竞争力提升策略，其输出格式可以包括但不限于html、doc、xml、pdf。此外，所述择校分析报告还可以包括当前申请者的竞争力分析结果相关的可视化数据图/表。所述择校分析报告还可以包括：按照被录取概率倒序排列的学校推荐列表、竞争力等级接近的学校的推荐列表以及推荐方案的对应相关服务提供商链接。具体生成的择校分析报告示意如图3所示，其中，竞争力提升策略部分仅以申请者选择申请甲学校为例生成对应的竞争力提升策略，在实际操作中针对申请者选择的各预选学校，均可生成对应的竞争力提升策略。
此外，所述择校分析与申请子系统800还包括推送模块820；所述推送模块820用于将当前申请者的所述择校分析报告发送至所述用户终端100。当前用户通过用户终端100接收生成择校分析报告，从而知悉自身的竞争力结果、被所选择的目标学校录取概率信息。
此外，所述择校分析与申请子系统800还包括申请模块830；所述申请模块830，用于实现对当前申请者相关的学校申请操作。例如，对当前申请者提交的申请资料进行初步审查；以及提交当前申请者的申请材料至官方入学申请系统；以及查询当前申请者的申请状态。
本实施例涉及的进行学校申请的个人竞争力智能评估系统将与学校申请高度相关的关键性非结构化数据及结构化数据均纳入至计分要素数据，并大量采集过往申请者的计分要素数据。基于大量过往申请者的计分要素数据，以及计分要素数据中的非结构化数据的量化，准确地分析出当前申请者的竞争力水平及对应不同实力水平等级学校的被录取概率，弥补了现有学习申请咨询工具无法结合申请者的非结构化数据进行个人学校申请竞争力分析的不足。同时，也为当前申请者提供了准确有效的竞争力提升策略、以及方便直观的择校分析报告和解决方案，从而为申请者提供了学校申请的充分直观、准确有效的个人竞争力评估信息，用以进行决策参考，有助于申请者更好地作出学校申请决策及增加申请成功率。本发明实施例实现了一站式智能在线学校申请咨询，与传统学校申请咨询（如留学咨询）需要高度的面对面或在线与咨询师交流不同，节省了申请者与咨询师交流的繁琐流程和高昂的沟通成本，帮助申请者实现了自助式的线上竞争力分析、找出不足并提供改进策略、在线申请及申请结果追踪等整套线上的学校申请准备全过程。
进一步地，如图1、图2所示，所述择校分析与申请子系统800包括择校分析模块810；
择校分析模块810，用于确定当前申请者对应预选学校的竞争力提升策略，对应的实现方式多种，例如：
在一些具体实施中，所述择校分析模块810，用于通过改变当前申请者的计分要素数据，计算对应预选学校的被录取概率的提升率；根据被录取概率的提升率，确定对应预选学校的最优可提升计分要素项及对应的竞争力提升策略。
即考察任一项计分要素数据的改变，对于预选学校的被录取概率的提升效果；若预选学校的被录取概率提升率较为明显，则改变的计分要素数据对应的计分要素项为较优或最优的可提升计分要素项。换言之，给当前申请者指明可以有效提高被预选学校录取概率的对应提升方向，从而制定对应的竞争力提升策略。
例如，领导活动活跃指数从当前的2分提升至3分，当前申请者被某一预选学校录取的概率从60%提升至65%；而发表论文数量（第一作者）从1篇提升至2篇时，被录取概率可以提升至85%。则显然，当前申请者更应该着重提升发表论文数量（第一作者），以提高被录取概率。其它的计分要素数据依次类推，获得对应被录取概率的提升率，从而找到最有利于当前申请者提高对应预选学校的被录取概率的提升计分要素项。
在另一些具体实施中，所述择校分析模块810，还用于根据预选学校的所述申请成功者模型及当前申请者的所述计分要素数据，确定当前申请者对应预选学校的申请成功者模型的薄弱计分要素项，以及确定对应的竞争力提升策略。
由于预选学校的申请成功者模型可以总体反映出预选学校的申请成功者的群体平均情况，包括各计分要素项的平均标准分（各预选学校的申请成功者的计分子要素分数的平均值）。将当前申请者的各计分要素数据与申请成功者模型的各计分要素项的平均标准分进行比较，更加简便地找出当前申请者的薄弱计分要素项，且计算方便准确。根据找出的薄弱计分要素项，向当前申请者提供建设性的竞争力提升策略（包括具体的优化建议），用以提高薄弱计分要素项的数据。例如，当前申请者的TOEFL成绩95分，明显低于某预选学校的申请成功者的TOEFL成绩平均标准分105分，则当前申请者应着重提高TOEFL成绩；对应地，生成的竞争力提升策略包含有提高TOEFL成绩的方法技巧和执行计划。
此外，还可以根据当前申请者的所述综合竞争力标准总分、所述竞争力等级及所述被录取概率，查找出潜在可以提升计分要素项，并相应地提供战略性改进建议，以使当前申请者提高申请成功率。
在一些具体实施中，根据当前申请者的所述综合竞争力标准总分、所述竞争力等级及所述被录取概率，生成当前申请者的竞争力提升策略。例如，比对当前申请者的竞争力等级对应的综合竞争力总分与更优一级竞争力等级对应的综合竞争力总分分值区间下限值，确定二者的分值差距，以及查找当前申请者的得分偏低的计分要素项，从而生成竞争力提升策略。举例来说，若当前申请者的课外活动计分要素得分偏低，则将课外活动计分要素项标记为需要改进的计分要素项。换言之，当前申请者通过用户终端获取该竞争力提升策略，并需要在课外活动计分要素项对应的各计分子要素项目进行提升和加强，以期提升竞争力水平。进一步地，对生成的竞争力提升策略进行可实现性分析，若分析结果为该竞争力改进方案中存在难以实现的改进项目，则对该改进项目进行标注以告知当前申请者。
在一些具体实施中，择校分析模块810还用于根据当前申请者的所述综合竞争力标准总分、所述竞争力等级及所述被录取概率，确定当前申请者的竞争力水平匹配的学校的实力水平等级；根据确定的学校的实力水平等级，查找处于该实力水平等级的学校，以及生成当前申请者被该实力水平等级的学校录取的竞争力提升策略。以及，查找优于上述实力水平等级的学校，以及被对应学校录取的竞争力提升策略。
在本实施例中，进行学校申请的个人竞争力智能评估系统向申请者提供多角度的具有建设性及可执行性的竞争力提升策略，从而有助于申请者了解自身的竞争力情况，并利用提供的竞争力提升策略作出针对性的改进及提升活动，有助于增加申请者的申请成功率。
进一步地，如图1、图2所示，所述数据建模与分析子系统700还包括模型优化模块730；其中，
所述模型优化模块730，用于基于预设测试数据，根据所述申请成功者模型确定竞争力预测结果，并获取所述竞争力评分子系统600基于所述预设测试数据确定的竞争力结果；将所述竞争力预测结果与所述竞争力结果进行比较，并根据得到的比较结果评估所述申请成功者模型的准确度，以获得准确度评估结果；根据所述准确度评估结果，优化所述申请成功者模型。
上述建模模型710建立申请成功者模型后，需要不断地对已建立的申请成功者模型进行评估，以评判模型的预测准确度，从而不断地进行模型的优化，以实现最佳的预测效果。其中，对申请成功者模型进行评估时，选用的评估数据为预测测试数据，优选为存储于数据库300中的过往申请者的计分要素数据。例如，将选取的随机若干个过往申请者的计分要素数据通过申请成功者模型进行运算和分析，得到竞争力预测结果（包括但不限于综合竞争力总分及竞争力等级）。然后，将选取的随机若干个过往申请者的计分要素数据通过竞争力评分子系统600中的竞争力总分模块610、竞争力等级模块620进行运算和分析，得到对应的竞争力结果（包括但不限于综合竞争力总分及竞争力等级）。将竞争力预测结果与竞争力结果进行比较，判断二者的差距；根据二者的差距判断结果，评判申请成功者模型的预测准确度。
更进一步地，若申请成功者模型的预测准确度较低，则对申请成功者模型进行模型优化或者建立新的申请成功者模型。此外，还可以根据模型的运行稳定性等其它评估指标对模型进行评估及优化。
这样，通过对建立的申请成功者模型不断地进行模型预测准确度评估，有助于模型的优化，从而保证申请成功者模型预测的准确度。
参照图4，本发明实施例提供一种进行学校申请的个人竞争力智能评估方法，包括以下步骤：
步骤S10，采集过往申请者的计分要素数据，以及接收当前申请者的计分要素数据，并将过往申请者的所述计分要素数据及当前申请者的所述计分要素数据存储至数据库；
其中，所述计分要素数据为申请者进行学校申请的关键考核数据，一般包括各类学校针对申请者的通用关键性考核数据。在一些实施例中，所述计分要素数据对应的计分要素项包括：背景资料、学术能力、课外活动、标准化考试及其它预设材料。需要注意的是，本发明各实施例中涉及的计分要素数据包括结构化数据及非结构化数据。结构化数据为可以直接量化的数据，如标准化考试的成绩分数；非结构化数据为无法直接量化的数据，即需要进行人为转化才可以成为量化数据，例如志愿者经历、工作经历等课外活动数据。
每项计分要素项包含不同的计分子要素项，可根据具体申请的学校类型或者评估需要进行自定义设置和增减，具体包括但不限于如下内容：
背景资料，至少包括：曾就读高中、曾就读大学、曾就读大学的院系学术声誉排名；
学术能力，包括但不限于高中GPA、大学GPA、大学专业GPA、年级/班级排名、荣誉课程(AP/Honor)、A-G科目（A为历史、社会科学，B为英语，C为数学，D为实验室科学，E为除英语之外的语言，F为视觉与表演艺术，G为大学预科选修课）、科研经历、发表论文情况；
课外活动，包括社团活动、体育活动、艺术才能、领导活动、预科情况、志愿者经历、工作经历；
标准化考试，包括AP/Honor成绩、TOEFL成绩、IELTS成绩、SAT成绩、SAT专业测试(SAT Subject)成绩、ACT成绩、GRE成绩、GRE专业测试成绩、GMAT成绩、LSAT成绩及其他语言和专业性测试成绩；
其他预设材料，包括但不限于：获奖情况、个人进步趋势、推荐信、自荐信。其中，获奖情况用于反映申请者获得奖项的重要程度；个人进步趋势用于反映申请者的学业/学术的进步和提升情况；推荐信用于反映外界对于申请者的个人评价；自荐信用于反映申请者对于自身的描述和自我评价。举例来说，若申请者在本科学习期间分别获得省级、国家级、国际级共三个学术奖项，且上述三个奖项的学术重要性依次递增，则在评价申请者的获奖情况时，给予不同学术重要性的学术奖项对应的分数，并依照奖项重要性设置相应权重，求取申请者在获奖情况要素的总分，从而对申请者的获奖情况进行量化计算及分析。
上述过往申请者包括历年曾经进行同一类学校申请的申请者，例如在对当前申请者进行海外留学申请的竞争力智能评估时，过往申请者应选择过往曾经进行过海外留学申请的申请者；在对当前申请者进行本国大学硕士申请的竞争力智能评估时，过往申请者应选择过往曾经进行过本国大学硕士申请的申请者。若某一过往申请者成功被某一学校录取，则将该过往申请者标记为“申请成功者”。其中，当前申请者的计分要素数据为当前申请者通过用户终端上传的计分要素数据，其中，用户终端可以是智能手机、平板电脑、笔记本电脑、台式电脑或者其它电子装置，通过用户终端，用户可以实现数据的相关处理，例如在用户终端的特定操作界面，当前申请者进行输入、上传、获取、检索并修改特定数据，例如计分要素数据，具体的计分要素数据含义在下文进一步说明。用户输入和修改的特定数据和自动采集的数据被存储在数据库中；此外，还可以根据用户的不同级别和角色设置来控制对存储数据的修改权限。
历年过往申请者的计分要素数据包括历年过往申请者的结构化计分要素数据和非结构化计分要素数据。过往申请者的计分要素数据中还包括学校公开数据。其中，学校公开数据包括对应学校公布的历年录取的结构化数据和非结构化数据，例如学校地理位置、学费情况、录取条件、院系学术声誉排名、历年录取人数、录取者的背景资料数据、学术能力数据、课外活动数据、标准化考试数据以及其他预设材料数据。具体地，运用大数据采集技术，大量采集并分析历年进行学校申请的申请者的计分要素数据。
此外，历年过往申请者的计分要素数据和当前申请者的计分要素数据存储于数据库；当需要调用计分要素数据时，调用数据库中存储的上述计分要素数据即可。
步骤S20，根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，确定当前申请者的综合竞争力标准总分及竞争力等级；
具体包括：步骤S21，根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，分别计算过往申请者各计分要素标准分、当前申请者的各计分要素标准分；
步骤S22，根据过往申请者各计分要素标准分、当前申请者的各计分要素标准分，计算当前申请者的所述综合竞争力标准总分；以及根据当前申请者的所述综合竞争力标准总分，确定当前申请者在预设群体中的排名情况；
下面，结合一优选实施例（以个人申请国外大学为例）进行说明。
（1）根据预设划分规则，将计分要素项中的各计分子要素项进行层次划分，一种优选的划分情况如表1所示。
表1
计分要素 计分 子要素 数据划分层次以及划分规则
第一层 第二层 第三层 第四层 第五层 第六层 第七层
背景资料 曾就读高中 排名 前10名 前30名 前50名 前70名 前100名 前150名 前150名以外
曾就读大学 排名 前10名 前30名 前50名 前70名 前100名 前150名 前150名以外
曾就读大学的院系学术声誉排名 前10名 前30名 前50名 前70名 前100名 前150名 前150名以外
学术能力 高中GPA 3.8 3.7 3.6 3.5 3.3 3.2 低于3.2
大学GPA 3.8 3.7 3.6 3.5 3.2 3 低于3
专业GPA 3.9 3.8 3.7 3.6 3.5 3.4 低于3.4
年级/班级排名 1% 3% 5% 7% 10% 12% 低于12%
荣誉课程(AP/Honor) 20 18 16 12 8 6 低于6
A-G 科目 56 54 52 50 48 46 低于46
科研经历 3 2 1 / / / /
发表论文数量（第一作者） 2 1 / / / / /
发表论文数量（第二作者） 3 2 1 / / / /
课外活动 社团活动 6 5 4 3 2 1 0
体育活动 6 5 4 3 2 1 0
艺术才能 3 3 3 / / / /
领导活动 2 2 2 / / / /
预科 1 1 1 / / / /
志愿者（小时） 500 400 300 200 100 50 低于50
工作经历 6 5 4 3 2 1 0
标准化考试 TOEFL成绩 118 115 110 102 94 80 低于80
IELTS成绩 9 8.5 8 7.5 7 6.5 低于6.5
SAT成绩 1500 1450 1400 1350 1250 1150 低于1150
SAT Subject成绩平均分 765 750 720 700 680 650 低于650
ACT成绩 33 32 30 29 26 23 低于23
GRE成绩 325 320 310 300 290 280 低于280
GRE Subject成绩 980 970 960 950 920 900 低于900
GMAT成绩 730 700 680 650 620 600 低于600
LSAT成绩 175 170 165 160 155 150 低于150
其他预设材料 获奖情况 / / / / / / /
个人进步趋势 / / / / / / /
举例来说，在教育背景项中，各项排名（包括曾就读高中排名、曾就读大学排名、曾就读大学的院系学术声誉排名）为相关教育权威评级机构公布的综合排名。若当前申请者的曾就读高中排名为前60名，则当前申请者的曾就读高中排名项属于第四层，其它的层次划分以此类推。又例如，GPA(Grade Point Average)即平均绩点，是用于评估学生成绩的一个指标，换算方法为把各个学科所得到的成绩换算为绩点，再按照各学科的学分比例进行加权求和，所得结果除以各学科学分总和即可得到平均绩点。表1中采用的是四分制GPA系统，包括第一层次（GPA等于3.8或以上）、第二层次（GPA等于3.7或以上）、第三层次（GPA等于3.6或以上）、第四层次（GPA等于3.5或以上）、第五层次（GPA等于3.3或以上）、第六层次（GPA等于3.2或以上）、第七层次（GPA小于3.2）。
将表1中每个子要素的每一层次数据赋予对应的层次参数和权重系数，通过加权求和的方式求取当前申请者某一计分要素项的各计分子要素分数Pi(e)，其计算公式为：
Pi(e)=∑ (每个计分子要素最大值*权重系数*层次参数)
然后，基于当前申请者某一计分要素项的各计分子要素分数Pi(e)，通过标准分数的常规求取方式计算当前申请者的各计分要素标准分Pi(es)。常规求取方式为现有技术，此处不再一一赘述。
然后，根据各计分要素分数Pi(e)，计算得到综合竞争力总分P(t)。具体计算方式如下：
将各计分要素分数Pi(e)分别乘以一个计分要素权重系数，分别得到各个乘积；将各个乘积相加求和，所得结果即为综合竞争力总分P(t)。具体的计分要素权重系数不作限制。这样，就可以得到当前申请者（或者过往申请者）的综合竞争力总分P(t)。
这样，就可以通过计算标准分数的方式得到当前申请者（或者过往申请者）的综合竞争力标准总分P(ts)。当前申请者的综合竞争力标准总分P(ts)可以客观地反映出当前申请者的竞争力水平。
然后，根据当前申请者的综合竞争力标准总分P(ts)，确定当前申请者在预设群体中的排名情况。
具体地，根据当前申请者的所述综合竞争力标准总分及纳入计算的申请者（过往申请者加上当前申请者）总数，可以很容易地获得当前申请者在过往申请者中的排名情况，从而精确地刻划出当前申请者相对于过往申请者的竞争力水平。此外，还可以通过获取与当前申请者同年进行学校申请的多数或者全部其它申请者的计分要素数据，并计算出当前申请者在当年申请的综合竞争力标准总分，从而获得当前申请者在当年申请者当中的排名情况，进而方便地确定出当前申请者相对于当年申请者的竞争力水平，提供当年申请的参考信息。
步骤S23，根据过往申请者的所述综合竞争力标准总分、当前申请者的所述综合竞争力标准总分，确定当前申请者的所述竞争力等级；
在一些具体实施中，竞争力等级模块620对所有过往申请者的综合竞争力标准总分P(ts)进行分值范围划分，分成不同的分值区间；其中，不同的分值区间对应不同的竞争力等级，例如较差、一般、较强、很强、极强五个等级。优选地，按照预设比例对所有过往申请者的综合竞争力标准总分P(ts)进行分值范围划分。在计算出当前申请者的综合竞争力标准总分P(ts)后，匹配出对应的分值范围，从而确定当前申请者的竞争力等级。
步骤S24，根据过往申请者的所述计分要素数据，确定各个学校的实力水平等级。
即在确定某一学校的实力水平等级时，从所有过往申请者中筛选出该学校历年的申请成功者，并按照上述方法依次求取该学校历年的申请成功者的各计分子要素分数Pi(e)、综合竞争力标准总分P(ts)及竞争力等级。在确定各个学校的实力水平等级时，优选地，先计算各个学校历年的申请成功者的综合竞争力标准总分P(ts)的平均值，然后对各个学校历年的申请成功者的综合竞争力标准总分P(ts)的平均值进行分值范围划分，分成不同的分值区间；其中，不同的分值区间对应不同的实力水平等级，例如A、B、C、D、E五个等级。需要说明的是，由于每个学校每年的申请成功者的计分要素数据可能会变化，因此在确定各所学校的实力水平等级时，需定期更新学校历年的申请成功者的计分要素数据，从而保证学校实力水平等级的准确性。
此外，除了上述确定各个学校的实力水平等级的方法以外，还可以引用权威机构发布的学校实力水平等级评定结果确定各个学校的实力水平等级。
步骤S30，根据预选学校的过往申请者的所述计分要素数据，建立预选学校的申请成功者模型；其中，预选学校为当前申请者选定的学校或者系统推荐的学校，即通过预选若干个学校进行学校申请的个人申请竞争力智能评估，最终得到与各预选学校对应的当前申请者的个人申请竞争力评估结果。
在一些具体实施中，根据过往申请者的计分要素数据及预设建模算法建立各个学校的申请成功者模型；其中，预设建模算法优选贝叶斯分类算法。即以采集的过往申请者者的计分要素数据为基础，将过往申请者的计分要素数据进行分类存储，再通过数据挖掘算法提取出每个过往申请者的每个计分要素属性和指标，以便进行数据分析和建模。所述数据挖掘算法包括但不限于以下算法：决策树分析算法、神经网络分析算法（neural network algorithm）、聚类分析算法、关联规则分析算法、逻辑斯谛回归分析算法（Logistic regression algorithm）。
在一些具体实施中，通过决策树分析算法将处于某一实力水平等级的学校的所有过往申请者的相关计分要素数据（包括计分要素属性及对应的指标数据）进行分析，从而得出所有因素对申请成功和申请失败的影响相关性。然后，通过聚类分析算法分析出申请成功者和申请失败者分别所具有的特有共同计分要素数据（包括计分要素属性及对应的指标数据）。将以上两类分析结果进行组合排序并建立若干个申请成功者模型，再通过对申请成功和失败者两类人的数据库中随机抽取样本进行取样，测试申请成功者模型的准确率，并将准确率最高的申请成功者模型作为用于对当前申请者进行被录取概率预测的申请成功者模型。
步骤S40，根据当前申请者的所述计分要素数据及预选学校的所述申请成功者模型，计算当前申请者对应预选学校的被录取概率；
在一些具体实施中，基于申请成功者模型，对当前申请者的计分要素数据进行运算，分析得出当前申请者对应不同预选学校的被录取概率。
步骤S50，确定当前申请者对应预选学校的竞争力提升策略；以及生成当前申请者的择校分析报告。
确定当前申请者对应预选学校的竞争力提升策略，目的在于：向申请者提供多角度的具有建设性及可执行性的竞争力提升策略，从而有助于申请者了解自身的竞争力情况，并利用提供的竞争力提升策略作出针对性的改进及提升活动，有助于增加申请者的申请成功率。
在一些具体实施中，当前申请者的所述择校分析报告至少包括当前申请者对应预选学校的被录取概率和竞争力提升策略。
需要注意的是，由于很多学校设有学生录取限额，以及很多学校对于特定国家的留学生也设有录取限额。因此，只提供当前申请者对应预选学校的被录取概率，并不能给当前申请者提供充分全面的参考信息。本发明与现有的传统人工咨询方式或者在线咨询系统进行咨询的方式的不同之处在于，本发明的智能评估系统不仅向当前申请者提供直观的个人综合排名以及在其所处竞争力等级的排名，还能通过刷选的方式向当前申请者提供在本国当年申请者中的排名。也即，本发明的智能评估系统可以基于当前申请者及其他申请者（包括过往申请者和当年申请者）的计分要素，计算出当前申请者的综合竞争力标准总分P(ts)、竞争力等级，并可以进一步地得到当前申请者在各类排名类别下的排名情况，从而使得当前申请者掌握充分且全面的自身竞争力水平信息，从而更有利于决策参考。因此，当前申请者的择校分析报告中还可以包括综合竞争力标准总分P(ts)、竞争力等级及相关的排名情况。
即所述择校分析报告中可包含当前申请者的自身竞争力水平信息（包括综合竞争力标准总分P(ts)、竞争力等级及相关的排名情况）、当前申请者对应若干个学校的被录取概率、竞争力提升策略，其输出格式可以包括但不限于html、doc、xml、pdf。此外，所述择校分析报告还可以包括当前申请者的竞争力分析结果相关的可视化数据图/表。所述择校分析报告还可以包括：按照被录取概率倒序排列的学校推荐列表、竞争力等级接近的学校的推荐列表以及推荐方案的对应相关服务提供商链接。具体生成的择校分析报告示意如图3所示，其中，竞争力提升策略部分仅以申请者选择申请甲学校为例生成对应的竞争力提升策略，在实际操作中针对申请者选择的各预选学校，均可生成对应的竞争力提升策略。
此外，还包括将当前申请者的所述择校分析报告发送至所述用户终端。当前用户通过用户终端接收生成择校分析报告，从而知悉自身的竞争力结果、被所选择的目标学校录取概率信息。
此外，还包括实现对当前申请者相关的学校申请操作。例如，对当前申请者提交的申请资料进行初步审查；以及提交当前申请者的申请材料至官方入学申请系统；以及查询当前申请者的申请状态。
本实施例将与学校申请高度相关的关键性非结构化数据及结构化数据均纳入至计分要素数据，并大量采集过往申请者的计分要素数据。基于大量过往申请者的计分要素数据，以及计分要素数据中的非结构化数据的量化，准确地分析出当前申请者的竞争力水平及对应不同实力水平等级学校的被录取概率，弥补了现有学校申请咨询工具无法结合申请者的非结构化数据进行学校申请的个人竞争力分析的不足。同时，也为当前申请者提供了准确有效的竞争力提升策略、以及方便直观的择校分析报告和解决方案，从而为当前申请者提供了学校申请的充分直观、准确有效的个人竞争力评估信息，用以进行决策参考，有助于申请者更好地作出学校申请决策及增加申请成功率。本发明实施例实现了一站式智能在线学校申请咨询，与传统学校申请咨询（如留学咨询）需要高度的面对面或在线与咨询师交流不同，节省了申请者与咨询师交流的繁琐流程和高昂的沟通成本，帮助申请者实现了自助式的线上竞争力分析、找出不足并提供改进策略、在线申请及申请结果追踪等整套线上的学校申请准备全过程。
进一步地，所述确定当前申请者对应预选学校的竞争力提升策略的步骤，具体包括：
步骤S51，通过改变当前申请者的计分要素数据，计算对应预选学校的被录取概率的提升率；根据被录取概率的提升率，确定对应预选学校的最优可提升计分要素项及对应的竞争力提升策略。
即考察任一项计分要素数据的改变，对于预选学校的被录取概率的提升效果；若预选学校的被录取概率提升率较为明显，则改变的计分要素数据对应的计分要素项为较优或最优的可提升计分要素项。换言之，给当前申请者指明可以有效提高被预选学校录取概率的对应提升方向，从而制定对应的竞争力提升策略。
例如，领导活动活跃指数从当前的2分提升至3分，当前申请者被某一预选学校录取的概率从60%提升至65%；而发表论文数量（第一作者）从1篇提升至2篇时，被录取概率可以提升至85%。则显然，当前申请者更应该着重提升发表论文数量（第一作者），以提高被录取概率。其它的计分要素数据依次类推，获得对应被录取概率的提升率，从而找到最有利于当前申请者提高对应预选学校的被录取概率的提升计分要素项。
进一步地，所述确定当前申请者对应预选学校的竞争力提升策略的步骤，具体还包括：
步骤S52，根据预选学校的所述申请成功者模型及当前申请者的所述计分要素数据，确定当前申请者对应预选学校的申请成功者模型的薄弱计分要素项，以及确定对应的竞争力提升策略。
由于预选学校的申请成功者模型可以总体反映出预选学校的申请成功者的群体平均情况，包括各计分要素项的平均标准分（各预选学校的申请成功者的计分子要素分数的平均值）。将当前申请者的各计分要素数据与申请成功者模型的各计分要素项的平均标准分进行比较，更加简便地找出当前申请者的薄弱计分要素项，且计算方便准确。根据找出的薄弱计分要素项，向当前申请者提供建设性的竞争力提升策略（包括具体的优化建议），用以提高薄弱计分要素项的数据。例如，当前申请者的TOEFL成绩95分，明显低于某预选学校的申请成功者的TOEFL成绩平均标准分105分，则当前申请者应着重提高TOEFL成绩；对应地，生成的竞争力提升策略包含有提高TOEFL成绩的方法技巧和执行计划。
此外，还可以根据当前申请者的所述综合竞争力标准总分、所述竞争力等级及所述被录取概率，查找出潜在可以提升计分要素项，并相应地提供战略性改进建议，以使当前申请者提高申请成功率。
在一些具体实施中，根据当前申请者的所述综合竞争力标准总分、所述竞争力等级及所述被录取概率，生成当前申请者的竞争力提升策略。例如，比对当前申请者的竞争力等级对应的综合竞争力总分与更优一级竞争力等级对应的综合竞争力总分分值区间下限值，确定二者的分值差距，以及查找当前申请者的得分偏低的计分要素，从而生成针对当前申请者实际竞争力情况的竞争力提升策略。举例来说，所述竞争力提升策略包括：若当前申请者的课外活动计分要素得分偏低，则将课外活动计分要素项标记为需要改进的计分要素项。换言之，当前申请者通过用户终端获取该竞争力提升策略，应对课外活动计分要素项对应的各计分子要素项目进行提升和加强，以期提升竞争力水平。进一步地，对生成的竞争力提升策略进行可实现性分析，若分析结果为该竞争力改进方案中存在难以实现的改进项目，则对该改进项目进行标注以告知当前申请者。
在一些具体实施中，根据当前申请者的所述综合竞争力标准总分、所述竞争力等级及所述被录取概率，确定当前申请者的竞争力水平匹配的学校的实力水平等级；根据确定的学校的实力水平等级，查找处于该实力水平等级的学校，以及生成当前申请者被该实力水平等级的学校录取的竞争力提升策略。以及，查找优于上述实力水平等级的学校，以及被对应学校录取的竞争力提升策略。
在本实施例中，通过向申请者提供多角度的具有建设性及可执行性的竞争力提升策略，从而有助于申请者了解自身的竞争力情况，并利用提供的竞争力提升策略作出针对性的改进及提升活动，有助于增加申请者的申请成功率。
进一步地，如图5所示，所述根据预选学校的过往申请者的所述计分要素数据，建立预选学校的申请成功者模型的步骤之后，还包括：
步骤S60，基于预设测试数据，根据所述申请成功者模型确定竞争力预测结果，并获取所述竞争力评分子系统基于所述预设测试数据确定的竞争力结果；
步骤S61，将所述竞争力预测结果与所述竞争力结果进行比较，并根据得到的比较结果评估所述申请成功者模型的准确度，以获得准确度评估结果；
步骤S62，根据所述准确度评估结果，优化所述申请成功者模型。
建立申请成功者模型后，需要不断地对已建立的申请成功者模型进行评估，以评判模型的预测准确度，从而不断地进行模型的优化，以实现最佳的预测效果。其中，对申请成功者模型进行评估时，选用的评估数据为预测测试数据，优选为存储于数据库中的过往申请者的计分要素数据。例如，将选取的随机若干个过往申请者的计分要素数据通过申请成功者模型进行运算和分析，得到竞争力预测结果（包括但不限于综合竞争力总分及竞争力等级）。然后，将选取的随机若干个过往申请者的计分要素数据进行运算和分析，得到对应的竞争力结果（包括但不限于综合竞争力总分及竞争力等级）。将竞争力预测结果与竞争力结果进行比较，判断二者的差距；根据二者的差距判断结果，评判申请成功者模型的预测准确度。
更进一步地，若申请成功者模型的预测准确度较低，则对申请成功者模型进行模型优化或者建立新的申请成功者模型。此外，还可以根据模型的运行稳定性等其它评估指标对模型进行评估及优化。
这样，通过对建立的申请成功者模型不断地进行模型预测准确度评估，有助于模型的优化，从而保证申请成功者模型预测的准确度。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者系统不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者系统所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者系统中还存在另外的相同要素。
上述本发明实施例序号仅仅为了描述，不代表实施例的优劣。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来。
以上仅为本发明的优选实施例，并非因此限制本发明的专利范围，凡是利用本发明说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本发明的专利保护范围内。

Claims (12)

  1. 一种进行学校申请的个人竞争力智能评估系统，包括：用户终端、连接网络、数据库，其特征在于，还包括：数据采集子系统、竞争力评分子系统、数据建模与分析子系统、择校分析与申请子系统；其中，
    所述数据采集子系统，用于采集过往申请者的计分要素数据，以及通过所述连接网络与所述用户终端连接，接收所述用户终端发送的当前申请者的计分要素数据，并将过往申请者的所述计分要素数据及当前申请者的所述计分要素数据存储至所述数据库；
    所述竞争力评分子系统，用于根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，确定当前申请者的综合竞争力标准总分及竞争力等级；
    所述数据建模与分析子系统包括建模模块、成功率分析模块；所述建模模块，用于根据预选学校的过往申请者的所述计分要素数据，建立预选学校的申请成功者模型；所述成功率分析模块，用于根据当前申请者的所述计分要素数据及预选学校的所述申请成功者模型，计算当前申请者对应预选学校的被录取概率；
    所述择校分析与申请子系统，用于确定当前申请者对应预选学校的竞争力提升策略；以及生成当前申请者的择校分析报告。
  2. 如权利要求1所述的进行学校申请的个人竞争力智能评估系统，其特征在于，所述计分要素数据对应的计分要素项包括：背景资料、学术能力、课外活动、标准化考试及其它预设材料。
  3. 如权利要求1所述的进行学校申请的个人竞争力智能评估系统，其特征在于，所述竞争力评分子系统包括竞争力总分模块、竞争力等级模块；其中，
    所述竞争力总分模块，用于根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，分别计算过往申请者各计分要素标准分、当前申请者的各计分要素标准分；根据过往申请者各计分要素标准分、当前申请者的各计分要素标准分，计算当前申请者的所述综合竞争力标准总分；
    以及根据当前申请者的所述综合竞争力标准总分，确定当前申请者在预设群体中的排名情况；
    所述竞争力等级模块，用于根据过往申请者的所述综合竞争力标准总分、当前申请者的所述综合竞争力标准总分，确定当前申请者的所述竞争力等级；以及根据过往申请者的所述计分要素数据，确定各个学校的实力水平等级。
  4. 如权利要求1所述的进行学校申请的个人竞争力智能评估系统，其特征在于，所述择校分析与申请子系统包括择校分析模块；
    所述择校分析模块，用于通过改变当前申请者的计分要素数据，计算对应预选学校的被录取概率的提升率；根据被录取概率的提升率，确定对应预选学校的最优可提升计分要素项及对应的竞争力提升策略。
  5. 如权利要求4所述的进行学校申请的个人竞争力智能评估系统，其特征在于，所述择校分析模块，还用于根据预选学校的所述申请成功者模型及当前申请者的所述计分要素数据，确定当前申请者对应预选学校的申请成功者模型的薄弱计分要素项，以及确定对应的竞争力提升策略。
  6. 如权利要求1、4或5任一所述的进行学校申请的个人竞争力智能评估系统，其特征在于，当前申请者的所述择校分析报告至少包括当前申请者对应预选学校的被录取概率、及/或竞争力提升策略、及/或综合竞争力标准总分、及/或竞争力等级。
  7. 如权利要求4所述的进行学校申请的个人竞争力智能评估系统，其特征在于，当前申请者的所述择校分析报告至少包括当前申请者对应预选学校的被录取概率、及/或竞争力提升策略、及/或综合竞争力标准总分、及/或竞争力等级。
  8. 如权利要求5所述的进行学校申请的个人竞争力智能评估系统，其特征在于，当前申请者的所述择校分析报告至少包括当前申请者对应预选学校的被录取概率、及/或竞争力提升策略、及/或综合竞争力标准总分、及/或竞争力等级。
  9. 如权利要求1所述的进行学校申请的个人竞争力智能评估系统，其特征在于，所述数据建模与分析子系统还包括模型优化模块；其中，
    所述模型优化模块，用于基于预设测试数据，根据所述申请成功者模型确定竞争力预测结果，并获取所述竞争力评分子系统基于所述预设测试数据确定的竞争力结果；将所述竞争力预测结果与所述竞争力结果进行比较，并根据得到的比较结果评估所述申请成功者模型的准确度，以获得准确度评估结果；根据所述准确度评估结果，优化所述申请成功者模型。
  10. 一种进行学校申请的个人竞争力智能评估方法，其特征在于，所述进行学校申请的个人竞争力智能评估方法包括：
    采集过往申请者的计分要素数据，以及接收当前申请者的计分要素数据，并将过往申请者的所述计分要素数据及当前申请者的所述计分要素数据存储至数据库；
    根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，确定当前申请者的综合竞争力标准总分及竞争力等级；
    根据预选学校的过往申请者的所述计分要素数据，建立预选学校的申请成功者模型；
    根据当前申请者的所述计分要素数据及预选学校的所述申请成功者模型，计算当前申请者对应预选学校的被录取概率；
    确定当前申请者对应预选学校的竞争力提升策略；以及生成当前申请者的择校分析报告。
  11. 如权利要求8所述的进行学校申请的个人竞争力智能评估方法，其特征在于，所述根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，确定当前申请者的综合竞争力标准总分及竞争力等级的步骤包括：
    根据过往申请者的所述计分要素数据、当前申请者的所述计分要素数据，分别计算过往申请者各计分要素标准分、当前申请者的各计分要素标准分；
    根据过往申请者各计分要素标准分、当前申请者的各计分要素标准分，计算当前申请者的所述综合竞争力标准总分；
    以及根据当前申请者的所述综合竞争力标准总分，确定当前申请者在预设群体中的排名情况；
    根据过往申请者的所述综合竞争力标准总分、当前申请者的所述综合竞争力标准总分，确定当前申请者的所述竞争力等级；
    以及根据过往申请者的所述计分要素数据，确定各个学校的实力水平等级。
  12. 如权利要求8所述的进行学校申请的个人竞争力智能评估方法，其特征在于，所述确定当前申请者对应预选学校的竞争力提升策略的步骤，具体包括：所述择校分析模块通过改变当前申请者的计分要素数据，计算对应预选学校的被录取概率的提升率；根据被录取概率的提升率，确定对应预选学校的最优可提升计分要素项及对应的竞争力提升策略；
    或者，根据预选学校的所述申请成功者模型及当前申请者的所述计分要素数据，确定当前申请者对应预选学校的申请成功者模型的薄弱计分要素项，以及确定对应的竞争力提升策略。
PCT/CN2019/074358 2018-02-12 2019-02-01 进行学校申请的个人竞争力智能评估系统及方法 WO2019154305A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810145377.4 2018-02-12
CN201810145377.4A CN108335047A (zh) 2018-02-12 2018-02-12 进行学校申请的个人竞争力智能评估系统及方法

Publications (1)

Publication Number Publication Date
WO2019154305A1 true WO2019154305A1 (zh) 2019-08-15

Family

ID=62929177

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/074358 WO2019154305A1 (zh) 2018-02-12 2019-02-01 进行学校申请的个人竞争力智能评估系统及方法

Country Status (2)

Country Link
CN (1) CN108335047A (zh)
WO (1) WO2019154305A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108335047A (zh) * 2018-02-12 2018-07-27 藕丝科技（深圳）有限公司 进行学校申请的个人竞争力智能评估系统及方法
CN109658296A (zh) * 2018-08-31 2019-04-19 北京沃达新创国际教育科技有限公司 一种智能留学服务平台
CN110688566A (zh) * 2019-09-06 2020-01-14 平安科技（深圳）有限公司 基于用户画像的数据推送方法、系统、设备及存储介质
CN113516372A (zh) * 2021-06-18 2021-10-19 广州启德教育科技有限公司 辅助留学选校的信息匹配方法及装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104574234A (zh) * 2014-07-17 2015-04-29 星腾科留学股份有限公司 大学申请机会最优化教育咨询系统
CN105184708A (zh) * 2015-08-14 2015-12-23 北京联校传奇信息科技有限公司 一种留学申请匹配方法及系统
CN106355530A (zh) * 2016-08-22 2017-01-25 深圳市乐博学坊咨询有限公司 一种高考志愿填报自动化分析与决策方法
CN106875028A (zh) * 2016-06-17 2017-06-20 江婕 一种辅助择校的信息筛选方法和系统
CN108335047A (zh) * 2018-02-12 2018-07-27 藕丝科技（深圳）有限公司 进行学校申请的个人竞争力智能评估系统及方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060069576A1 (en) * 2004-09-28 2006-03-30 Waldorf Gregory L Method and system for identifying candidate colleges for prospective college students
US20150006423A1 (en) * 2013-06-28 2015-01-01 ThinkTank Learning Inc. Individualized education consulting system and method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104574234A (zh) * 2014-07-17 2015-04-29 星腾科留学股份有限公司 大学申请机会最优化教育咨询系统
CN105184708A (zh) * 2015-08-14 2015-12-23 北京联校传奇信息科技有限公司 一种留学申请匹配方法及系统
CN106875028A (zh) * 2016-06-17 2017-06-20 江婕 一种辅助择校的信息筛选方法和系统
CN106355530A (zh) * 2016-08-22 2017-01-25 深圳市乐博学坊咨询有限公司 一种高考志愿填报自动化分析与决策方法
CN108335047A (zh) * 2018-02-12 2018-07-27 藕丝科技（深圳）有限公司 进行学校申请的个人竞争力智能评估系统及方法

Also Published As

Publication number Publication date
CN108335047A (zh) 2018-07-27

Similar Documents

Publication Publication Date Title
WO2019154305A1 (zh) 进行学校申请的个人竞争力智能评估系统及方法
WO2019059717A1 (ko) 기업 분석 기반 연계 서비스 매칭 플랫폼 제공 시스템
US20080071746A1 (en) Method For Interactive Employment Searching, Rating, And Selecting of Employment Listing
Maleyeff et al. Six Sigma and introductory statistics education
WO2019050074A1 (ko) 클라우드 기반의 디지털 문항 저작 솔루션 제공 및 유통 서비스 플랫폼이 가능한 학습시스템과, 그 제어방법
WO2021033827A1 (ko) 딥러닝 모듈을 이용한 발달장애 개선 시스템 및 방법
WO2014115963A1 (ko) 재능경력중심학점 인정 학사관리시스템 및 학사관리방법
WO2017116215A1 (ko) 평가지표 자율제안에 의한 연구개발과제 선정 시스템 및 방법
WO2024096203A1 (ko) 통증 환자를 위한 인지 행동 치료 제공 방법 및 시스템
WO2020022611A1 (ko) 재능경력중심 학점인정 학사관리 시스템 및 방법과 이를 이용하는 재능기부은행 서비스 제공 시스템
WO2020001031A1 (zh) 学校及专业推荐方法及系统
Kiger et al. A new method for group decision making and its application in medical trainee selection
CN106920429A (zh) 一种信息处理方法及装置
Mallery et al. A stochastic model of consensus reaching in committee decisions for faculty advancement, promotion and tenure: does diversity matter?
Lan et al. [Retracted] Evaluation System of Music Art Instructional Quality Based on Convolutional Neural Networks and Big Data Analysis
WO2023113553A1 (ko) 클라우드 서버 기반의 인지 또는 언어 재활 훈련을 위한 컨텐츠 제공 방법
Jantzen Do economic principal applicants work in their intended regulated occupation? Introducing the national household survey and immigration landing file linkage database
Chang et al. The features and marketability of certificates for occupational safety and health management in Taiwan
KR20070091843A (ko) 휴대 단말을 통한 코스 코디네이터 시스템 및 방법
WO2022045844A1 (ko) 파이낸셜 플래너 모집을 위한 정보 제공 방법 및 이를 이용한 전자 장치
CN115222564A (zh) 一种在线学习平台课程智能推荐方法
Shilpa et al. A Measurement tool to study the e-readiness of extension service providers of agricultural research and development institutions in Telangana
WO2024080422A1 (ko) 비대면 채용 서비스를 제공하는 인공지능 기반의 전문 인적자원 플랫폼 서비스 방법
Veledar et al. A Bibliometric Overview of Business Ethics
KR20160065493A (ko) 한국사 학습을 위한 문제 출제 및 테스트 시스템과 이를 이용한 한국사 능력 레벨 테스트 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19750856

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 13/01/2021)

122 Ep: pct application non-entry in european phase

Ref document number: 19750856

Country of ref document: EP

Kind code of ref document: A1