技术领域technical field
本发明涉及互联网问卷技术领域，确切地说涉及一种调查问卷自动生成技术。The invention relates to the technical field of Internet questionnaires, and more specifically, to an automatic generation technology of questionnaires.
背景技术Background technique
调查问卷通常由一系列的问题、调查项目、备选答案或者填写说明组成。主要目的是向被调查者收集项目相关统计数据。调查问卷作为收集研究资料的工具，可以获取到大量的统计数据，对这些信息进行深入的分析，可验证研究假设并对所研究的问题作出科学的解释和说明。在实际调查中，要想通过问卷搜集高效度、高信度的研究数据，问卷设计是保证调查结果有效的关键点之一。设计问卷时设计人员往往需要考虑问卷结构的确定、试题的编写与排序、答案选项及量表的选择、试题表述与调查对象的关系等问题，同时无法根据当前收集到的信息快速对问卷进行调整。因此，有效的自动化问卷生成及自动化采集调查数据可极大地带来便利，减少相关人员工作量的同时降低人工失误的可能性。Questionnaires usually consist of a series of questions, survey items, alternative answers, or instructions for filling in. The main purpose is to collect project-related statistics from respondents. Questionnaire, as a tool for collecting research data, can obtain a large amount of statistical data, conduct in-depth analysis of this information, verify research hypotheses and give scientific explanations and explanations to the research issues. In the actual survey, in order to collect high-efficiency and high-reliability research data through questionnaires, the design of the questionnaire is one of the key points to ensure the effectiveness of the survey results. When designing a questionnaire, designers often need to consider issues such as the determination of the questionnaire structure, the writing and sorting of test questions, the selection of answer options and scales, the relationship between the expression of the test questions and the survey objects, etc. At the same time, it is impossible to quickly adjust the questionnaire according to the currently collected information. . Therefore, effective automatic questionnaire generation and automatic collection of survey data can greatly bring convenience, reduce the workload of relevant personnel and reduce the possibility of manual errors.
由于问卷调查项目通常涉及到许多的考察方面，于是需要一个系统的指标体系来指导问卷调查工作。成熟的问卷调查项目涉及到的指标数量较为庞大，并且各个指标都需要有一定的数据支持。因此问卷试题应该覆盖所有的指标，同时需要体现不同指标的不同重要程度。Because the questionnaire survey project usually involves many aspects of investigation, a systematic index system is needed to guide the questionnaire survey work. A mature questionnaire survey project involves a relatively large number of indicators, and each indicator needs to be supported by certain data. Therefore, the questionnaire questions should cover all indicators, and at the same time need to reflect the different degrees of importance of different indicators.
受调查人群在做问卷时耐心程度有限。当问卷题目过多时，受调查者放弃调查的几率大大增加，这样不仅浪费了资源，还使得获取到的有效数据减少。如何根据不同人群的耐心程度来设计调查问卷，已经成为现有技术中亟待解决的问题之一。The respondents have limited patience when doing the questionnaires. When there are too many questions in the questionnaire, the probability of respondents giving up the survey is greatly increased, which not only wastes resources, but also reduces the effective data obtained. How to design questionnaires according to the degree of patience of different groups of people has become one of the problems to be solved urgently in the prior art.
为了解决大量数据需求与受调查人群耐心程度有限以及预算有限三者间的矛盾，现在出现了一些调查问卷自动生成技术。In order to solve the contradiction between the demand for a large amount of data and the limited patience of the surveyed population and the limited budget, some automatic questionnaire generation technologies have emerged.
如现有技术中，如公开号为CN107194743A，公开日期为2017年9月22的，名称为“一种网络调查问卷生成方法和装置”的中国发明专利文献，公开了一种网络调查问卷生成方法和装置，用以在尽可能少的问题中收集更多的数据，提高调查问卷的数据回收效率和网络资源利用率。网络问卷调查方法，包括：接收用户设备提交的接受网络问卷调查邀请的响应消息；针对问题库中包含的可选问题，根据预设的评估参数确定该可选问题对应的权重；并按照权重由大至小的顺序选择预设数量的可选问题；根据所述问题库中包含的必选问题和选择出的可选问题生成网络调查问卷；向所述用户设备返回生成的网络调查问卷。As in the prior art, such as the publication number CN107194743A, the publication date is September 22, 2017, and the Chinese invention patent document entitled "A method and device for generating a network questionnaire" discloses a method for generating a network questionnaire And the device is used to collect more data in as few questions as possible, improve the data recovery efficiency of the questionnaire and the utilization rate of network resources. A network questionnaire survey method includes: receiving a response message submitted by a user equipment for accepting an invitation to a network questionnaire survey; for optional questions included in a question base, determining a weight corresponding to the optional question according to preset evaluation parameters; Selecting a preset number of optional questions in descending order; generating a network questionnaire according to the mandatory questions and the selected optional questions contained in the question base; returning the generated network questionnaire to the user equipment.
然而这种技术方案仅针对在线问答的调查，并不能实现线下问卷的自动生成，并无法解决现在问卷调查中容错率、完成率、有效率等方面的问题，并且调查受众也局限于在线用户，广度不足。However, this technical solution is only for online question and answer surveys, and cannot automatically generate offline questionnaires, and cannot solve the problems of fault tolerance, completion rate, and efficiency in current questionnaire surveys, and the survey audience is also limited to online users. , the breadth is insufficient.
发明内容SUMMARY OF THE INVENTION
本发明的目的在于在有限预算下，使用问卷生成算法自动生成与需求相适应、并根据受调查人群的耐心程度动态调整问卷结构，尽可能多地收集调查数据的调查问卷自动生成技术。The purpose of the present invention is to use a questionnaire generation algorithm to automatically generate a questionnaire automatic generation technology that adapts to the demand, dynamically adjusts the questionnaire structure according to the patient degree of the investigated population, and collects as much survey data as possible under a limited budget.
本发明的目的是通过以下技术方案实现的：The purpose of this invention is to realize through the following technical solutions:
一种调查问卷自动生成方法，其特征在于：包括题库生成步骤、卷类型构成步骤、问卷题数确定步骤和问卷生成步骤；A method for automatically generating a questionnaire, which is characterized in that it comprises a step of generating a question bank, a step of forming a questionnaire type, a step of determining the number of questions in a questionnaire, and a step of generating a questionnaire;
所述题库生成步骤，问卷调查工作通常需要一个系统的指标体系作为指导，该指标体系中最底层的指标之和即为指标总数，且每个底层指标与一个或多个试题相关联，这些试题共同构成问卷题库；In the question bank generation step, a systematic index system is usually required for the questionnaire survey work. The sum of the bottom-level indexes in the index system is the total number of indexes, and each bottom-level index is associated with one or more test questions. Together, they form a question bank of questionnaires;
所述卷类型构成步骤，根据某次问卷调查的预算来确定问卷总数S，且问卷总数S与问卷调查预算之间呈正相关；按照问卷需求设定问卷类型的数量L；In the step of forming the questionnaire types, the total number of questionnaires S is determined according to the budget of a certain questionnaire survey, and there is a positive correlation between the total number of questionnaires S and the questionnaire survey budget; the number L of questionnaire types is set according to the questionnaire requirements;
所述问卷题数确定步骤，确定问卷类型的数量L、设定问卷的试题总数M，所述问卷的试题总数至少不少于指标总数、至多等于整个题库的试题总数；根据问卷的试题总数M和问卷类型数量L计算每份问卷的试题数量X＝M/L；即每份问卷的试题数量与问卷类型数的乘积即为试题总数；The step of determining the number of questions in the questionnaire is to determine the number L of the types of the questionnaire, and the total number of questions M of the set questionnaire. Calculate the number of questions in each questionnaire X=M/L with the number of questionnaire types L; that is, the product of the number of questions in each questionnaire and the number of questionnaire types is the total number of questions;
所述问卷生成步骤，按照问卷类型构成步骤和问卷题数确定步骤中确定的每份问卷的问卷类型和试题数量m，调用问卷生成算法生成问卷，所述问卷生成算法通过对题库中的已经编号的试题进行处理得到一个试题编号序列，并根据试题编号序列从题库中抽取试题组成问卷。In the questionnaire generation step, according to the questionnaire type and the number of questions m of each questionnaire determined in the questionnaire type composition step and the questionnaire number determination step, the questionnaire generation algorithm is invoked to generate the questionnaire, and the questionnaire generation algorithm The test questions are processed to obtain a test question number sequence, and the test questions are extracted from the question bank according to the test question number sequence to form a questionnaire.
所述问卷数量构成步骤中，问卷的总数S，问卷类型数量L；问卷的总数S确定后根据需求指定问卷类型数量L，同一类型下问卷的数量记为Ci，问卷的总数S为所有问卷类型下问卷数量的总和，即其中i∈N+；问卷类型数量L增加则同一类型下的问卷数量Ci减少；由上述公式可确定问卷类型数及各类型下问卷数量的关系。In the step of forming the number of questionnaires, the total number of questionnaires S, the number of questionnaire types L; after the total number of questionnaires S is determined, the number of questionnaire types L is specified according to the needs, the number of questionnaires under the same type is recorded as Ci , the total number of questionnaires S is all questionnaires The sum of the number of questionnaires under the type, i.e. where i∈N+ ; as the number of questionnaire types L increases, the number of questionnaires Ci under the same type decreases; the relationship between the number of questionnaire types and the number of questionnaires under each type can be determined by the above formula.
所述问卷题数确定步骤中，设定问卷的试题总数M，指标总数Z，整个题库的试题总数T，每份问卷的试题数量X；且其中X∈N+，N+为整数，由此可以确定每份问卷的试题数量X。In the step of determining the number of questions in the questionnaire, set the total number of questions M of the questionnaire, the total number of indicators Z, the total number of questions T of the entire question bank, and the number of questions of each questionnaire X; and Among them, X∈N+ , N+ is an integer, from which the number of test questions X of each questionnaire can be determined.
所述问卷生成步骤中，所述问卷生成算法包括以下步骤：In the step of generating the questionnaire, the algorithm for generating the questionnaire includes the following steps:
步骤1，为题库中的试题按照与底层指标的关联性进行编号；Step 1, number the test questions in the question bank according to the correlation with the underlying indicators;
步骤2，生成试题编号矩阵，将每个指标关联的试题编号集合作为试题编号矩阵中的一列，若矩阵行数不相等，则数量较少的列缺少的部分用所在列的试题编号填补，得到一个列为指标总数且行相等的试题编号矩阵；Step 2: Generate a test question number matrix, and use the set of test question numbers associated with each index as a column in the test question number matrix. If the number of rows in the matrix is not equal, the missing part of the column with a smaller number is filled with the test question number of the column, and the result is obtained. A matrix of question numbers with equal rows and columns for the total number of indicators;
步骤3，先将每一列的编号在该列内进行随机排列，得到一个新的试题编号矩阵，然后顺序将每一行的编号依次放入一个新的集合中，得到可用于生成问卷的试题编号序列；Step 3: Randomly arrange the numbers of each column in this column to obtain a new test item number matrix, and then sequentially put the numbers of each row into a new set to obtain the test item number sequence that can be used to generate the questionnaire. ;
步骤4，按照试题编号序列生成问卷，根据步骤3中编号序列顺序抽取试题，即每X个试题作为一份新的问卷，以此循环直到取到设定的试题总数M，如果最后一份问卷试题数量不不足X时，再从序列重新开始顺序选取编号补齐，这样生成的问卷可以覆盖所有指标；每份问卷生成的同时还需要为每份问卷编号并对每份试卷的试题编号进行记录，同时加上时间戳，便于后续的问卷回收及调查结果录入。Step 4, generate the questionnaire according to the number sequence of the test questions, and extract the test questions according to the sequence of the number sequence in step 3, that is, every X test questions are used as a new questionnaire, and this cycle will be repeated until the set total number of test questions M is obtained. If the last questionnaire When the number of test questions is not less than X, the number is selected from the sequence to make up, so that the generated questionnaire can cover all indicators; each questionnaire needs to be numbered and the test question number of each test paper should be recorded at the same time. , and add a timestamp to facilitate subsequent questionnaire recovery and input of survey results.
而根据使用状态，为题库中的试题按照与底层指标的关联性进行编号的过程可以是在题库生成步骤中即开始进行的，先为底层指标进行编号，然后为每个底层指标关联的试题进行编号，通过如前缀+后缀等方式对试题完成编号。According to the usage status, the process of numbering the test questions in the question bank according to the correlation with the underlying indicators can be started in the step of generating the question bank. Numbering, complete the numbering of the test questions by means such as prefix + suffix.
所述步骤1中，为指标编号并标记为Zj，且j∈N+；每一个指标Zj都有与之对应的题库，按照指标对应的题库为试题编号，如指标Z1关联的试题编号集合为{Z(1，1),Z(1,2),Z(1,3),...,Z(1，n)}。In the step 1, the index is numbered and marked as Zj , and j∈N+ ; each index Zj has a corresponding question bank, and the question bank corresponding to the index is the test question number, such as the test question associated with the index Z1 The numbered set is {Z(1,1) ,Z(1,2) ,Z(1,3) ,...,Z(1,n) }.
还包括问卷回收步骤，回收调查问卷后，利用扫描技术自动获取问卷编号及问卷的结果，将这些信息送入系统，系统自动为问卷评分并纳入统计。It also includes a questionnaire recovery step. After the questionnaire is recovered, the questionnaire number and the results of the questionnaire are automatically obtained by scanning technology, and these information are sent to the system, and the system automatically scores the questionnaire and includes it into statistics.
还包括问卷评价步骤，系统为问卷回收步骤中回收的问卷进行评分，收集调查数据，答题数量过少的问卷视为无效数据；如果未收集到目标数量的数据，则根据数据情况调整问卷的结构，减少问卷试题数量，进行数据的补充。优选地，如一份问卷试题中受试者答题的数目小于该份试题数目的70％，则视该份问卷结果无效，系统自动记录无效问卷编号和答题数目；统计完成后，若某类型问卷的数据量不足，则根据无效问卷的情况对每份问卷试题数量X进行调整，生成新的补充问卷。最后回收的新问卷结果作为补充数据，以同样的方法录入系统。It also includes a questionnaire evaluation step. The system scores the questionnaires recovered in the questionnaire recovery step and collects survey data. Questionnaires with too few answers are regarded as invalid data; if the target number of data is not collected, the structure of the questionnaire will be adjusted according to the data situation. , reduce the number of questionnaires and supplement the data. Preferably, if the number of questions answered by subjects in a questionnaire is less than 70% of the number of questions in the questionnaire, the result of the questionnaire will be regarded as invalid, and the system will automatically record the number of the invalid questionnaire and the number of answers; If the amount of data is insufficient, the number of questions X in each questionnaire will be adjusted according to the situation of invalid questionnaires, and a new supplementary questionnaire will be generated. The results of the new questionnaire collected at the end were used as supplementary data and entered into the system in the same way.
与现有技术相比，本发明具有以下优点：Compared with the prior art, the present invention has the following advantages:
本发明提出的一种自动化生成问卷的方法，利用数学的方法生成试题编号序列，再根据该序列自动生成问卷，既丰富了问卷的类型，也为问卷设计提供了极大的便利；问卷生成过程中生成随机的试题编号序列，该序列同时覆盖了调查问卷的指标及所有的题库。并且利用该试题序列生成问卷避免了问卷中同一指标的重复出现，解决了调查问卷设计过程中无法兼顾指标与题库的问题，使得生成的问卷更加科学与高效。The method for automatically generating a questionnaire proposed by the present invention uses a mathematical method to generate a sequence of test question numbers, and then automatically generates a questionnaire according to the sequence, which not only enriches the types of the questionnaire, but also provides great convenience for the design of the questionnaire; the process of generating the questionnaire A random numbering sequence of test questions is generated in the system, and the sequence covers the indicators of the questionnaire and all the question banks at the same time. And the use of the test question sequence to generate the questionnaire avoids the repetition of the same index in the questionnaire, solves the problem that the index and the question bank cannot be taken into account in the process of questionnaire design, and makes the generated questionnaire more scientific and efficient.
本发明提出的自动化生成问卷的方法后续还会对数量未达标的数据进行补充收集，即根据已获得的数据推断受调查者的耐心程度，调整每份问卷的试题数量后进行数据补充，使得获取到的数据数量充足并且有效性高。The method for automatically generating questionnaires proposed by the present invention will subsequently supplement the collection of data that does not meet the standard, that is, according to the obtained data to infer the degree of patience of the respondent, adjust the number of test questions in each questionnaire, and then supplement the data, so that the obtained data can be supplemented. The amount of data received is sufficient and the validity is high.
附图说明Description of drawings
本发明的前述和下文具体描述在结合以下附图阅读时变得更清楚，其中：The foregoing and following detailed description of the present invention will become more apparent when read in conjunction with the following accompanying drawings, in which:
图1为本发明调查问卷自动生成方法一种优选方案的逻辑示意图；Fig. 1 is the logical schematic diagram of a kind of preferred scheme of the automatic generation method of questionnaire of the present invention;
图2为本发明问卷生成算法一种优选方案的逻辑示意图。Fig. 2 is a schematic diagram of a preferred solution of the questionnaire generation algorithm of the present invention.
具体实施方式Detailed ways
下面通过几个具体的实施例来进一步说明实现本发明目的技术方案，需要说明的是，本发明要求保护的技术方案包括但不限于以下实施例。The technical solutions for realizing the object of the present invention are further described below through several specific embodiments. It should be noted that the technical solutions claimed in the present invention include but are not limited to the following embodiments.
实施例1Example 1
作为本发明一种最基本的实施方案，本实施例公开了一种调查问卷自动生成方法，如图1所示，包括题库生成步骤、卷类型构成步骤、问卷题数确定步骤和问卷生成步骤；As one of the most basic implementations of the present invention, this embodiment discloses a method for automatically generating a questionnaire, as shown in Figure 1, including a step of generating a question bank, a step of forming a questionnaire type, a step of determining the number of questions in a questionnaire, and a step of generating a questionnaire;
所述题库生成步骤，问卷调查工作通常需要一个系统的指标体系作为指导，该指标体系中最底层的指标之和即为指标总数，且每个底层指标与一个或多个试题相关联，这些试题共同构成问卷题库；In the question bank generation step, a systematic index system is usually required for the questionnaire survey work. The sum of the bottom-level indexes in the index system is the total number of indexes, and each bottom-level index is associated with one or more test questions. Together, they form a question bank of questionnaires;
所述卷类型构成步骤，根据某次问卷调查的预算来确定问卷总数S，且问卷总数S与问卷调查预算之间呈正相关；按照问卷需求设定问卷类型的数量L；In the step of forming the questionnaire types, the total number of questionnaires S is determined according to the budget of a certain questionnaire survey, and there is a positive correlation between the total number of questionnaires S and the questionnaire survey budget; the number L of questionnaire types is set according to the questionnaire requirements;
所述问卷题数确定步骤，确定问卷类型的数量L、设定问卷的试题总数M，所述问卷的试题总数至少不少于指标总数、至多等于整个题库的试题总数；根据问卷的试题总数M和问卷类型数量L计算每份问卷的试题数量X＝M/L；即每份问卷的试题数量与问卷类型数的乘积即为试题总数；The step of determining the number of questions in the questionnaire is to determine the number L of the types of the questionnaire, and the total number of questions M of the set questionnaire. Calculate the number of questions in each questionnaire X=M/L with the number of questionnaire types L; that is, the product of the number of questions in each questionnaire and the number of questionnaire types is the total number of questions;
所述问卷生成步骤，按照问卷类型构成步骤和问卷题数确定步骤中确定的每份问卷的问卷类型和试题数量m，调用问卷生成算法生成问卷，所述问卷生成算法通过对题库中的已经编号的试题进行处理得到一个试题编号序列，并根据试题编号序列从题库中抽取试题组成问卷。In the questionnaire generation step, according to the questionnaire type and the number of questions m of each questionnaire determined in the questionnaire type composition step and the questionnaire number determination step, the questionnaire generation algorithm is invoked to generate the questionnaire, and the questionnaire generation algorithm The test questions are processed to obtain a test question number sequence, and the test questions are extracted from the question bank according to the test question number sequence to form a questionnaire.
本实施例的自动化生成问卷的方法，利用数学的方法生成试题编号序列，再根据该序列自动生成问卷，既丰富了问卷的类型，也为问卷设计提供了极大的便利；问卷生成过程中生成随机的试题编号序列，该序列同时覆盖了调查问卷的指标及所有的题库。并且利用该试题序列生成问卷避免了问卷中同一指标的重复出现，解决了调查问卷设计过程中无法兼顾指标与题库的问题，使得生成的问卷更加科学与高效。The method for automatically generating a questionnaire in this embodiment uses a mathematical method to generate a sequence of test question numbers, and then automatically generates a questionnaire according to the sequence, which not only enriches the types of questionnaires, but also provides great convenience for questionnaire design; A random test question numbering sequence, which covers both the indicators of the questionnaire and all the question banks. And the use of the test question sequence to generate the questionnaire avoids the repetition of the same index in the questionnaire, solves the problem that the index and the question bank cannot be taken into account in the process of questionnaire design, and makes the generated questionnaire more scientific and efficient.
实施例2Example 2
作为本发明一种优选的实施方案，在实施例1的技术方案基础上，本实施例进一步公开了所述问卷数量构成步骤中，问卷的总数S，问卷类型数量L；问卷的总数S确定后根据需求指定问卷类型数量L，同一类型下问卷的数量记为Ci，问卷的总数S为所有问卷类型下问卷数量的总和，即其中i∈N+；问卷类型数量L增加则同一类型下的问卷数量Ci减少；由上述公式可确定问卷类型数及各类型下问卷数量的关系。As a preferred embodiment of the present invention, based on the technical solution of Embodiment 1, this embodiment further discloses that in the step of forming the number of questionnaires, the total number of questionnaires S and the number of types of questionnaires L; after the total number of questionnaires S is determined Specify the number of questionnaire types L according to the needs, the number of questionnaires under the same type is denoted as Ci , and the total number of questionnaires S is the sum of the number of questionnaires under all questionnaire types, namely where i∈N+ ; as the number of questionnaire types L increases, the number of questionnaires Ci under the same type decreases; the relationship between the number of questionnaire types and the number of questionnaires under each type can be determined by the above formula.
所述问卷题数确定步骤中，设定问卷的试题总数M，指标总数Z，整个题库的试题总数T，每份问卷的试题数量X；且其中X∈N+，N+为整数，由此可以确定每份问卷的试题数量X。In the step of determining the number of questions in the questionnaire, set the total number of questions M of the questionnaire, the total number of indicators Z, the total number of questions T of the entire question bank, and the number of questions of each questionnaire X; and Among them, X∈N+ , N+ is an integer, from which the number of test questions X of each questionnaire can be determined.
还包括问卷评价步骤，系统为问卷回收步骤中回收的问卷进行评分，收集调查数据，答题数量过少的问卷视为无效数据；如果未收集到目标数量的数据，则根据数据情况调整问卷的结构，减少问卷试题数量，进行数据的补充。优选地，如一份问卷试题中受试者答题的数目小于该份试题数目的70％，则视该份问卷结果无效，系统自动记录无效问卷编号和答题数目；统计完成后，若某类型问卷的数据量不足，则根据无效问卷的情况对每份问卷试题数量X进行调整，生成新的补充问卷。最后回收的新问卷结果作为补充数据，以同样的方法录入系统；对数量未达标的数据进行补充收集，即根据已获得的数据推断受调查者的耐心程度，调整每份问卷的试题数量后进行数据补充，使得获取到的数据数量充足并且有效性高。It also includes a questionnaire evaluation step. The system scores the questionnaires recovered in the questionnaire recovery step and collects survey data. Questionnaires with too few answers are regarded as invalid data; if the target number of data is not collected, the structure of the questionnaire will be adjusted according to the data situation. , reduce the number of questionnaires and supplement the data. Preferably, if the number of questions answered by subjects in a questionnaire is less than 70% of the number of questions in the questionnaire, the result of the questionnaire will be regarded as invalid, and the system will automatically record the number of the invalid questionnaire and the number of answers; If the amount of data is insufficient, the number of questions X in each questionnaire will be adjusted according to the situation of invalid questionnaires, and a new supplementary questionnaire will be generated. The results of the new questionnaires collected at the end are used as supplementary data, and are entered into the system in the same way; supplementary collection of data that does not meet the standard is carried out, that is, the degree of patience of the respondents is inferred based on the obtained data, and the number of test questions for each questionnaire is adjusted. The data is supplemented, so that the amount of data obtained is sufficient and the validity is high.
进一步的，如图2所示，所述问卷生成步骤中，所述问卷生成算法包括以下步骤：Further, as shown in Figure 2, in the step of generating the questionnaire, the algorithm for generating the questionnaire includes the following steps:
步骤1，为题库中的试题按照与底层指标的关联性进行编号；Step 1, number the test questions in the question bank according to the correlation with the underlying indicators;
步骤2，生成试题编号矩阵，将每个指标关联的试题编号集合作为试题编号矩阵中的一列，若矩阵行数不相等，则数量较少的列缺少的部分用所在列的试题编号填补，得到一个列为指标总数且行相等的试题编号矩阵；Step 2: Generate a test question number matrix, and use the set of test question numbers associated with each index as a column in the test question number matrix. If the number of rows in the matrix is not equal, the missing part of the column with a smaller number is filled with the test question number of the column, and the result is obtained. A matrix of question numbers with equal rows and columns for the total number of indicators;
步骤3，先将每一列的编号在该列内进行随机排列，得到一个新的试题编号矩阵，然后顺序将每一行的编号依次放入一个新的集合中，得到可用于生成问卷的试题编号序列；Step 3: Randomly arrange the numbers of each column in this column to obtain a new test item number matrix, and then sequentially put the numbers of each row into a new set to obtain the test item number sequence that can be used to generate the questionnaire. ;
步骤4，按照试题编号序列生成问卷，根据步骤3中编号序列顺序抽取试题，即每X个试题作为一份新的问卷，以此循环直到取到设定的试题总数M，如果最后一份问卷试题数量不不足X时，再从序列重新开始顺序选取编号补齐，这样生成的问卷可以覆盖所有指标；每份问卷生成的同时还需要为每份问卷编号并对每份试卷的试题编号进行记录，同时加上时间戳，便于后续的问卷回收及调查结果录入。Step 4, generate the questionnaire according to the number sequence of the test questions, and extract the test questions according to the sequence of the number sequence in step 3, that is, every X test questions are used as a new questionnaire, and this cycle will be repeated until the set total number of test questions M is obtained. If the last questionnaire When the number of test questions is not less than X, the number is selected from the sequence to make up, so that the generated questionnaire can cover all indicators; each questionnaire needs to be numbered and the test question number of each test paper should be recorded at the same time. , and add a timestamp to facilitate subsequent questionnaire recovery and input of survey results.
而根据使用状态，为题库中的试题按照与底层指标的关联性进行编号的过程可以是在题库生成步骤中即开始进行的，先为底层指标进行编号，然后为每个底层指标关联的试题进行编号，通过如前缀+后缀等方式对试题完成编号。According to the usage status, the process of numbering the test questions in the question bank according to the correlation with the underlying indicators can be started in the step of generating the question bank. Numbering, complete the numbering of the test questions by means such as prefix + suffix.
所述步骤1中，为指标编号并标记为Zj，且j∈N+；每一个指标Zj都有与之对应的题库，按照指标对应的题库为试题编号，如指标Z1关联的试题编号集合为{Z(1，1),Z(1,2),Z(1,3),...,Z(1，n)}。In the step 1, the index is numbered and marked as Zj , and j∈N+ ; each index Zj has a corresponding question bank, and the question bank corresponding to the index is the test question number, such as the test question associated with the index Z1 The numbered set is {Z(1,1) ,Z(1,2) ,Z(1,3) ,...,Z(1,n) }.
还包括问卷回收步骤，回收调查问卷后，利用扫描技术自动获取问卷编号及问卷的结果，将这些信息送入系统，系统自动为问卷评分并纳入统计。It also includes a questionnaire recovery step. After the questionnaire is recovered, the questionnaire number and the results of the questionnaire are automatically obtained by scanning technology, and these information are sent to the system, and the system automatically scores the questionnaire and includes it into statistics.
实施例3Example 3
作为本发明一种优选的实施方案，本实施例公开了一种调查问卷自动生成方法，包括以下步骤：As a preferred embodiment of the present invention, the present embodiment discloses a method for automatically generating a questionnaire, comprising the following steps:
步骤1、根据问卷调查的预算计算问卷的总数，然后根据根据问卷总数确定问卷类型数；同一类型的问卷试题相同，不同类型的问卷试题不同，从而确定问卷数量的构成。Step 1. Calculate the total number of questionnaires according to the budget of the questionnaire survey, and then determine the number of questionnaire types according to the total number of questionnaires; the same type of questionnaire has the same questions, and different types of questionnaires have different questions, so as to determine the composition of the number of questionnaires.
问卷调查中预算有限，需要根据预算来确定问卷的数量。设问卷预算与问卷总数(用S表示)的关系为正相关。当S确定时，可根据需求指定问卷类型数(用L表示)。其中，不同类型的问卷试题不同，而相同类型的问卷试题相同，同一类型下问卷的数量记为Ci。问卷总数为所有类型下问卷数量的总和，即问卷类型数L增加，同一类型下的问卷数量Ci则会减少。由上述公式可确定问卷类型数及各类型下问卷数量的关系The budget in the questionnaire survey is limited, and the number of questionnaires needs to be determined according to the budget. The relationship between the questionnaire budget and the total number of questionnaires (represented by S) is assumed to be positive. When S is determined, the number of questionnaire types (represented by L) can be specified according to requirements. Among them, different types of questionnaires have different questions, but the same type of questionnaires have the same questions, and the number of questionnaires under the same type is recorded as Ci . The total number of questionnaires is the sum of the number of questionnaires under all types, namely As the number of questionnaire types L increases, the number of questionnaires Ci under the same type decreases. The above formula can determine the relationship between the number of questionnaire types and the number of questionnaires under each type
步骤2、调查问卷试题总量至少需要覆盖整个指标体系，在此前提下，用户也可指定试题总数；因此在确定好问卷类型数量后，需要根据问卷试题的总数来计算每份问卷的试题数量。Step 2. The total number of questions in the questionnaire should at least cover the entire index system. On this premise, the user can also specify the total number of questions; therefore, after determining the number of questionnaire types, the number of questions for each questionnaire needs to be calculated based on the total number of questions. .
当确定问卷类型数后，用户可指定试题总数(记为M)，试题总数至少覆盖整个指标体系(指标总数记为Z)，至多覆盖整个题库(题库中试题总数记为T)。然后根据用户指定的试题总数M来计算每份问卷的试题数量(记为X)。每份问卷试题数N与问卷类型数L的乘积为试题总数M。并且每份问卷的试题数量X需要满足公式由此可以确定每份问卷的试题数量X。After determining the number of questionnaire types, the user can specify the total number of questions (denoted as M), the total number of questions covers at least the entire index system (the total number of indicators is denoted as Z), and at most covers the entire question bank (the total number of questions in the question bank is denoted as T). Then, according to the total number of test questions M specified by the user, the number of test questions for each questionnaire (denoted as X) is calculated. The product of the number of questions N of each questionnaire and the number of types of questionnaires L is the total number of questions M. And the number of questions X of each questionnaire needs to satisfy the formula From this, the number of questions X for each questionnaire can be determined.
步骤3、生成问卷时需按照确定好的问卷类型数量及每份问卷的试题数量来进行；调用问卷生成算法，算法首先要为所有指标编号，同时为指标关联题库中的试题编号；将这些已经编号的试题进行处理，最终得到一个可行的试题编号序列；最后顺序从试题编号序列中抽取试题，组成不同类型的问卷。记录问卷编号以便系统自动处理调查结果。Step 3. When generating the questionnaire, it should be done according to the determined number of questionnaire types and the number of test questions for each questionnaire; when calling the questionnaire generation algorithm, the algorithm must first number all the indicators, and at the same time, number the test questions in the index-related question bank; The numbered test questions are processed, and finally a feasible test question numbering sequence is obtained; the final sequence is to extract test questions from the test question numbering sequence to form different types of questionnaires. Record the questionnaire number so that the system can automatically process the survey results.
问卷生成算法具体步骤如下：The specific steps of the questionnaire generation algorithm are as follows:
1、为指标及试题编号。算法为所有指标编号并标记为Zj(j∈N+)。对于每一个指标Zj，都有与之对应的题库。按照指标题库为试题编号，如指标Zj关联的试题编号集合为{Z11，Z12，Z13，…，Z1n}。1. Number of indicators and test questions. The algorithm numbers and labels all indicators as Zj (j∈N+ ). For each indicator Zj , there is a corresponding question bank. The test questions are numbered according to the index library, for example, the set of test question numbers associated with the index Zj is {Z11 , Z12 , Z13 , ..., Z1n }.
2、生成试题编号矩阵。将每个指标关联的试题编号集合作为试题编号矩阵中的一列，若矩阵行数不相等，则数量较少的列缺少的部分用所在列的试题编号填补，最终得到一个列为指标总数且行相等的试题编号矩阵。2. Generate the test question number matrix. The set of question numbers associated with each indicator is used as a column in the question number matrix. If the number of rows in the matrix is not equal, the missing part of the column with a smaller number is filled with the question number of the column, and finally a column with the total number of indicators and equal rows is obtained. The test item number matrix.
3、得到新的试题编号序列。算法先将每一列的编号在该列内进行随机排列，得到一个新的试题编号矩阵。然后顺序将每一行的编号依次放入一个新的集合中。得到可用于生成问卷的试题编号序列。3. Get a new test question numbering sequence. The algorithm first randomly arranges the numbers of each column in the column to obtain a new test item number matrix. Then sequentially put the number of each row into a new set in turn. Get a sequence of question numbers that can be used to generate the questionnaire.
4、按照试题编号序列生成问卷。算法根据上述编号序列顺序抽取试题，即每N个试题作为一份新的问卷，以此循环直到取到用户指定的试题总数M。如果最后一份问卷试题数量不足X时，再从序列重新开始顺序选取编号补齐。这样生成的问卷可以覆盖所有指标。每份问卷生成的同时还需要为每份问卷编号并对每份试卷的试题编号进行记录，同时加上时间戳。便于后续的问卷回收及调查结果录入。4. The questionnaire is generated according to the sequence of test question numbers. The algorithm selects the test questions according to the above-mentioned number sequence, that is, every N test questions is used as a new questionnaire, and this cycle is repeated until the total number of test questions M specified by the user is obtained. If the number of questions in the last questionnaire is less than X, the number will be selected from the sequence to make up. The questionnaire thus generated can cover all indicators. When each questionnaire is generated, it is also necessary to number each questionnaire and record the question number of each test paper, and add a time stamp at the same time. It is convenient for follow-up questionnaire recovery and input of survey results.
进一步的，为评价问卷质量和不断完善题库和算法，还包括以下步骤：Further, in order to evaluate the quality of the questionnaire and continuously improve the question bank and algorithm, the following steps are also included:
步骤4、回收调查问卷，采用扫描的形式自动获取问卷的编号。根据问卷的编号自动从数据库中调出该问卷相关信息，然后按照试题编号记录受试者答案。系统会自动根据受试者答案进行评分并记录数据。Step 4. Recover the questionnaire, and automatically obtain the number of the questionnaire in the form of scanning. According to the number of the questionnaire, the relevant information of the questionnaire is automatically retrieved from the database, and then the answers of the subjects are recorded according to the number of the test question. The system automatically scores and records data based on the subject's answers.
步骤5、无效问卷的情况可以在一定程度上反映受调查人群的耐心程度。若某份问卷中受试者答题的数目小于该份试题数目的70％，则视该份问卷结果无效，系统自动记录无效问卷编号和答题数目。统计完成后，若某类型问卷的数据量不足，则根据无效问卷的情况对每份问卷试题数量X进行调整，生成新的补充问卷。最后回收的新问卷结果作为补充数据，以同样的方法录入系统。Step 5. The situation of invalid questionnaires can reflect the patience of the surveyed population to a certain extent. If the number of subjects answered in a certain questionnaire is less than 70% of the number of the test questions, the result of the questionnaire is regarded as invalid, and the system automatically records the invalid questionnaire number and the number of answers. After the statistics are completed, if the data volume of a certain type of questionnaire is insufficient, the number of questions X of each questionnaire will be adjusted according to the situation of invalid questionnaires, and a new supplementary questionnaire will be generated. The results of the new questionnaire collected at the end were used as supplementary data and entered into the system in the same way.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910404782.8ACN110147953B (en) | 2019-05-16 | 2019-05-16 | A Method for Automatically Generating Questionnaire |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910404782.8ACN110147953B (en) | 2019-05-16 | 2019-05-16 | A Method for Automatically Generating Questionnaire |
| Publication Number | Publication Date |
|---|---|
| CN110147953Atrue CN110147953A (en) | 2019-08-20 |
| CN110147953B CN110147953B (en) | 2023-01-10 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910404782.8AActiveCN110147953B (en) | 2019-05-16 | 2019-05-16 | A Method for Automatically Generating Questionnaire |
| Country | Link |
|---|---|
| CN (1) | CN110147953B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113032574A (en)* | 2021-05-27 | 2021-06-25 | 明品云(北京)数据科技有限公司 | Questionnaire configuration method, system, equipment and medium based on keywords |
| CN114896367A (en)* | 2022-05-12 | 2022-08-12 | 中国工商银行股份有限公司 | Test paper generation method and device and electronic equipment |
| CN115630613A (en)* | 2022-12-19 | 2023-01-20 | 长沙冉星信息科技有限公司 | Automatic coding system and method for evaluation problems in questionnaire survey |
| CN118114890A (en)* | 2024-04-30 | 2024-05-31 | 佛山市城市规划设计研究院有限公司 | Urban physical examination questionnaire investigation method and related equipment thereof |
| CN119539995A (en)* | 2024-11-21 | 2025-02-28 | 江苏达海智能系统股份有限公司 | A smart building data management method and system based on data collection |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060122465A1 (en)* | 2002-06-13 | 2006-06-08 | Philippe Bastien | Methods and systems for generating diagnostic algorithms based on questionnaires |
| CN101017540A (en)* | 2007-01-24 | 2007-08-15 | 华为技术有限公司 | Intelligent questionnaire system, questionnaire development method and device thereof |
| CN101315680A (en)* | 2007-05-31 | 2008-12-03 | 中国科学院自动化研究所 | Group Opinion Qualitative Analysis Tool and Implementation Method Based on Automatic Questionnaire |
| CN104008453A (en)* | 2014-05-29 | 2014-08-27 | 启秀科技（北京）有限公司 | Vocational ability evaluation simulation system |
| CN105117867A (en)* | 2015-10-09 | 2015-12-02 | 张娟 | Design and implementation of enterprise questionnaire system |
| CN105825270A (en)* | 2016-03-03 | 2016-08-03 | 云南电网有限责任公司教育培训评价中心 | Automatic test paper generation method for test questions in question bank through genetic algorithm |
| CN106803123A (en)* | 2016-12-20 | 2017-06-06 | 广州中软信息技术有限公司 | A kind of automatic volume group method for online exam |
| US20170178163A1 (en)* | 2015-12-17 | 2017-06-22 | B3Intelligence Ltd. | Method and system for generating a split questionnaire |
| CN107194743A (en)* | 2017-06-19 | 2017-09-22 | 腾讯科技（深圳）有限公司 | A kind of network surveying questionnaire generation method and device |
| CN108428152A (en)* | 2018-03-12 | 2018-08-21 | 平安科技（深圳）有限公司 | questionnaire generation method, server and computer readable storage medium |
| CN109050103A (en)* | 2018-11-06 | 2018-12-21 | 电子科技大学中山学院 | Questionnaire statistics record board based on big data analysis |
| CN110245213A (en)* | 2019-05-08 | 2019-09-17 | 平安科技（深圳）有限公司 | Questionnaire generation method, device, equipment and storage medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060122465A1 (en)* | 2002-06-13 | 2006-06-08 | Philippe Bastien | Methods and systems for generating diagnostic algorithms based on questionnaires |
| CN101017540A (en)* | 2007-01-24 | 2007-08-15 | 华为技术有限公司 | Intelligent questionnaire system, questionnaire development method and device thereof |
| CN101315680A (en)* | 2007-05-31 | 2008-12-03 | 中国科学院自动化研究所 | Group Opinion Qualitative Analysis Tool and Implementation Method Based on Automatic Questionnaire |
| CN104008453A (en)* | 2014-05-29 | 2014-08-27 | 启秀科技（北京）有限公司 | Vocational ability evaluation simulation system |
| CN105117867A (en)* | 2015-10-09 | 2015-12-02 | 张娟 | Design and implementation of enterprise questionnaire system |
| US20170178163A1 (en)* | 2015-12-17 | 2017-06-22 | B3Intelligence Ltd. | Method and system for generating a split questionnaire |
| CN105825270A (en)* | 2016-03-03 | 2016-08-03 | 云南电网有限责任公司教育培训评价中心 | Automatic test paper generation method for test questions in question bank through genetic algorithm |
| CN106803123A (en)* | 2016-12-20 | 2017-06-06 | 广州中软信息技术有限公司 | A kind of automatic volume group method for online exam |
| CN107194743A (en)* | 2017-06-19 | 2017-09-22 | 腾讯科技（深圳）有限公司 | A kind of network surveying questionnaire generation method and device |
| CN108428152A (en)* | 2018-03-12 | 2018-08-21 | 平安科技（深圳）有限公司 | questionnaire generation method, server and computer readable storage medium |
| CN109050103A (en)* | 2018-11-06 | 2018-12-21 | 电子科技大学中山学院 | Questionnaire statistics record board based on big data analysis |
| CN110245213A (en)* | 2019-05-08 | 2019-09-17 | 平安科技（深圳）有限公司 | Questionnaire generation method, device, equipment and storage medium |
| Title |
|---|
| 周文君: "一种高效的随机组卷算法的设计", 《电脑与电信》* |
| 李晓英: "基于K-means聚类的调查问卷动态赋权统计方法", 《统计与决策》* |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113032574A (en)* | 2021-05-27 | 2021-06-25 | 明品云(北京)数据科技有限公司 | Questionnaire configuration method, system, equipment and medium based on keywords |
| CN113032574B (en)* | 2021-05-27 | 2021-10-08 | 明品云(北京)数据科技有限公司 | Questionnaire configuration method, system, equipment and medium based on keywords |
| CN114896367A (en)* | 2022-05-12 | 2022-08-12 | 中国工商银行股份有限公司 | Test paper generation method and device and electronic equipment |
| CN115630613A (en)* | 2022-12-19 | 2023-01-20 | 长沙冉星信息科技有限公司 | Automatic coding system and method for evaluation problems in questionnaire survey |
| CN118114890A (en)* | 2024-04-30 | 2024-05-31 | 佛山市城市规划设计研究院有限公司 | Urban physical examination questionnaire investigation method and related equipment thereof |
| CN119539995A (en)* | 2024-11-21 | 2025-02-28 | 江苏达海智能系统股份有限公司 | A smart building data management method and system based on data collection |
| Publication number | Publication date |
|---|---|
| CN110147953B (en) | 2023-01-10 |
| Publication | Publication Date | Title |
|---|---|---|
| CN110147953A (en) | A method of automatic generation of questionnaires | |
| CN106815784B (en) | Based on the examination system for improving ant group algorithm intelligent Auto-generating Test Paper | |
| CN105138624B (en) | A kind of personalized recommendation method based on online course user data | |
| US8666302B2 (en) | System and method for education including community-sourced data and community interactions | |
| JP2004503877A5 (en) | ||
| CN108460627A (en) | Marketing activity scheme method for pushing, device, computer equipment and storage medium | |
| CN115346654A (en) | Intelligent service system based on internet | |
| CN113256252A (en) | B/S architecture-based test system | |
| CN110020208A (en) | Position recommender system | |
| KR102502734B1 (en) | Patient satisfaction evaluation service provision system for hospitals for Voice of Patient management | |
| CN117743695B (en) | Vocational training course recommendation method and device, electronic equipment and storage medium | |
| Győrffy et al. | Factors influencing the scientific performance of Momentum grant holders: an evaluation of the first 117 research groups | |
| CN116012205A (en) | An intelligent training management method and system for elderly nurses | |
| Ahmed et al. | Evaluating the effectiveness of author-count based metrics in measuring scientific contributions | |
| CN115630613B (en) | Automatic coding system and method for evaluation problems in questionnaire survey | |
| Powers et al. | Career research productivity patterns of marketing academicians | |
| CN118397886A (en) | Interactive data supervision method and system based on MVC framework | |
| KR102009713B1 (en) | Online marketing education system | |
| Nandhi et al. | User experience evaluation using the cognitive walkthrough method | |
| Kaiser et al. | A New Era of Online Search? A Large-Scale Study of User Behavior and Personal Preferences during Practical Search Tasks with Generative AI versus Traditional Search Engines | |
| Rahma et al. | Student argumentation skill in physics learning: Bibliometric analysis | |
| Shrivastava et al. | Social data analysis: Cyber recruitment analysis spam detection over twitter dataset using SVM & ARIMA model | |
| Gopalakrishnan et al. | TEXT MINING TRANSPORTATION RESEARCH GRANT BIG DATA: KNOWLEDGE EXTRACTION AND PREDICTIVE MODELING USING FAST NEURAL NETS. | |
| CN115687788A (en) | Intelligent business opportunity recommendation method and system | |
| Rimland | Do we do it good well? A bibliographic essay on the evaluation of reference effectiveness |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |