WO2015074612A1 - 一种在线统计群体满意度的方法 - Google Patents

一种在线统计群体满意度的方法 Download PDF

Info

Publication number
WO2015074612A1
WO2015074612A1 PCT/CN2014/091988 CN2014091988W WO2015074612A1 WO 2015074612 A1 WO2015074612 A1 WO 2015074612A1 CN 2014091988 W CN2014091988 W CN 2014091988W WO 2015074612 A1 WO2015074612 A1 WO 2015074612A1
Authority
WO
WIPO (PCT)
Prior art keywords
satisfaction
assignment
group
option
curve
Prior art date
Application number
PCT/CN2014/091988
Other languages
English (en)
French (fr)
Inventor
韩李宾
李剑
徐烽
Original Assignee
韩李宾
李剑
徐烽
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 韩李宾, 李剑, 徐烽 filed Critical 韩李宾
Publication of WO2015074612A1 publication Critical patent/WO2015074612A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201Market modelling; Market analysis; Collecting market data

Definitions

  • the invention relates to a method for online statistical group satisfaction.
  • the method of statistical group satisfaction is through the "one person, one vote” voting method.
  • A, B, and C are three alternatives to a, b, and c. They can assign certain scores to a, b, and c according to the degree of preference, in order to express different degrees of preference. .
  • the true social preference order is: a>b>c
  • the personal social preference is completely eliminated, and the order of social preference is:
  • the social order preference is (a>b).
  • the object of the present invention is to provide a simple and accurate method for interpreting and extracting the online statistical group satisfaction of the group wishes, thereby solving disputes and making decisions online fairly through the network.
  • the reason why the present invention is not discussed is that the present invention not only requires the participants to sort but also requires the participants to quantify the options.
  • a method for online statistical group satisfaction comprising the following steps:
  • the statistical initiator determines in which group the statistics are performed, so that the statistical initiator completes the satisfaction statistics in the designated group;
  • the group's most satisfactory assignment option is counted: after the satisfaction assignment is completed, the group assignment of each assignment option is counted, and the assignment option with the largest group satisfaction is obtained after comparison, and the highest satisfaction value is assigned to all assignment options.
  • the option is the most satisfying statistical result for the group.
  • the assignment options are divided into a single option and a collective option; if an option can use the continuous change of the value in the option to change the content of the option, the option is called a collective option; if an option cannot be utilized If the value of the option changes continuously to change the content of the option, the option is called a single option.
  • Assignment options can be various combinations of singular and collective options.
  • N is preferably 10.
  • the participating evaluator directly gives a satisfaction value to the option.
  • the participating assigners For the collective option, the participating assigners only need to assign satisfaction to the local part of the assignment option, and then can set the satisfaction assignment in the entire definition domain of the assignment option by setting the satisfaction assignment function; thus, the individual satisfaction can be synthesized.
  • the satisfaction assignment function must be a unimodal bounded function; the representative application is an asymmetric normal distribution density function.
  • Participating assignors use three-point assignment or multi-point assignment to assign values. Whenever an assignment method is used, it must contain the most satisfactory selection point and at least one evaluation point around the point.
  • the corresponding unimodal assignment curve is solved by the undetermined coefficient method; that is, the satisfaction assignment of the participating assignees in the entire definition domain of the collective option is obtained. If the participating assignor does not assign a value to an option, the group assignment is taken as his personal assignment.
  • the specific algorithm for using the three-point assignment method can use the algorithm for simulating the satisfaction curve of each participating assignor using an asymmetric normal distribution density function, as follows:
  • x is the value of the assignment option and f(x) is the satisfaction value.
  • the participating assignees determine the three points on the satisfaction curve by the three-point assignment method. If the most full The value of the option is m, the satisfaction is k when the value of the option is 0, and the satisfaction is v when the maximum value is max.
  • the specific algorithm when using the multipoint assignment method can use the "algorithm of simulating the satisfaction curve of each participating assignor with a piecewise function" as follows:
  • the first derivative of f 0 (x 0 ) y 0
  • f 0 (x 0 ) is equal to the first derivative of f 1 (x 0 )
  • f m+1 (x m ) y m
  • the first derivative of m+1 (x m ) is equal to the first derivative of f m (x m ).
  • the coefficient of the function is determined by the undetermined coefficient method. After the coefficients are obtained, the individual satisfaction function f(x) of all segments can be obtained, and the individual satisfaction evaluation curve is obtained, so that the personal satisfaction of the value of any option can be determined.
  • the participation evaluator assigns an average value of the satisfaction value of the assignment option as a unit of the singularity option; assuming two individual satisfaction values, y 1 , y 2 , then they The composite group assignment is If there are n individual satisfaction assignments, y 1 , y 2 , ... y n , then the group assignments they synthesize are For the aggregation option, the satisfaction curve of each participating assignor obtained by calculation is used to synthesize the group satisfaction curve. The highest point of the group satisfaction curve is the collective option group assignment; suppose there are two individual satisfaction curves.
  • the statistical initiator shall express the disputed event and designate the party;
  • the statistical initiator determines in which group the statistics are performed, so that the statistical initiator completes the statistics in the designated group; the group determines the event or the party.
  • the group's reward and punishment measures that can be performed by the party are pre-listed as an evaluation option; it can be set by the statistical initiator, or the statistical initiator can set the opinion after consulting the group;
  • the group's most satisfactory assignment option is counted: after the satisfaction assignment is over, the group assignment of the assignment option is counted, and the option of the group satisfaction is obtained after comparison.
  • the participating assigners are all randomly selected.
  • the advantage of the random sampling is that the network water army and the fan group can avoid the malicious evaluation of the population.
  • the sampling statistics have a 99% probability that the sampling statistics will reach 90 points or more. It is said that the sampling method has 99% reliability and achieves accuracy of 90 points or more.
  • Table 1 The specific contents of Table 1 are as follows:
  • the group in the present invention is reproducible for the decision to resolve the dispute, and the conclusion of the statistic is unique.
  • the decision sponsor will be the statistical initiator to describe the event. Depending on the type of event, when the party is required to be designated, the party should be designated;
  • the invention has the advantages that the above-mentioned online statistical group satisfaction method has the characteristics of easy operation and accurate statistics, and solves the defect that the traditional method cannot truly express the will of the group; and since the invention does not need to be introduced like the traditional dispute resolution mode, A third-party operator mediates or arbitrates parties to online disputes, which greatly reduces the cost of dispute resolution and improves the efficiency of dispute resolution.
  • the invention has wide application, and can not only apply proposal investigation, public opinion survey, group decision, but also can be applied to dispute management events such as community management, online shopping disputes, forum disputes, and Q group disputes.
  • FIG. 1 is a flow chart of a method for online statistical group satisfaction according to the present invention.
  • An online statistical group satisfaction method is applied online to solve online shopping disputes.
  • An online statistical group satisfaction method for online dispute resolution including the following steps:
  • the statistical sponsor determines that the group performing the statistics is all members of the trading website, including buyers and sellers;
  • the assignment options can be various, such as rewarding and penalizing a certain amount of money, prohibiting the transaction of a certain number of days; the website agrees to reward and reward a certain amount of money as an assignment. Representative of the option;
  • the satisfaction range is [-10, +10] real number
  • the participation of the assignee is a satisfaction of a penalty of up to 10 for a fine of 2,000 yuan
  • a satisfaction of a fine of 1,000 yuan is [-10, +10] real number
  • the degree is 5, the satisfaction of a fine of 3,000 yuan is 6, and thus 3 points are obtained;
  • the satisfaction coordinate system is created, the X axis in the coordinate system represents the reward and punishment value (unit: yuan), and the Y axis represents the individual to x Satisfaction of reward and punishment, according to the assignment result of the participating assigners, the asymmetric normal distribution density function is used to synthesize the satisfaction value of the participating assignors in all the domains, and the individual's satisfaction curve for the penalty money is obtained;
  • the group's most satisfactory assignment option is calculated: according to the obtained satisfaction score curve of each participating assignor on the reward and punishment money, the satisfaction evaluation curve of the group on the reward and punishment money is synthesized.
  • n is 60.
  • the group satisfaction curve achieves the highest satisfaction value on the penalty of 1800 yuan.
  • the point of highest satisfaction is a fine of 1800 yuan, which is the most satisfactory assignment option for the group. It is also the result of the group ruling of the online dispute.

Landscapes

  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本发明公开了一种在线统计群体满意度的方法。本发明将满意度作为度量单位衡量人的意愿,实际上本发明给出了一种提取群体意愿的简易方法,群体满意度统计以数值形式进行。首先,统计发起人确定需统计的事件并进行表述;然后,在群体内随机抽取参与赋值者,由其对相关的赋值选项给出满意度赋值,并通过数学方法合成出每个参与赋值者在赋值选项全部定义域里的满意度赋值,同时合成出群体的满意度曲线;最后,对比各个赋值选项的赋值结果,满意度最高的赋值选项即为群体最满意的赋值选项。由于采用上述在线统计群体满意度的方法,具有易操作、统计精确的特点,解决了传统一人一票的投票方法不能真正表达群体意愿的缺陷。

Description

一种在线统计群体满意度的方法 技术领域
本发明涉及一种在线统计群体满意度的方法。
背景技术
目前统计群体满意度的方法,是通过“一人一票”的投票方式。
目前的投票统计方法在提取群体满意度时的缺陷:“一人一票”在候选项目达到“三”项和至少三名投票者时,无法正确萃取民意,无法正确表达民意。
在“一人一票”的投票中,当候选项目是“三”的情况下,由于每位投票者手中的选票只有一张,若将选票投向其中一项,则对另外两项的偏好程度就被抹杀了,无法表达出对另外两项的偏好信息,只有把他们统统归为“不喜欢”,这显然是荒谬的。根据“阿罗不可能定理”即使允许投票者对所有选项进行排序时,也会引出悖论。
比如下列状况中,假设甲、乙、丙三人,面对a、b、c三个备选方案,他们可以根据偏好程度分别赋予a、b、c一定的分值,以表达偏好程度的不同。
甲(a>c>b)
65 25 10
乙(b>a>c)
50 30 20
丙(c>b>a)
40 35 25
合计:
a获得65+30+25=120
b获得10+50+35=95
c获得25+20+40=85
真实的社会偏好次序为:a>b>c
120 95 85
而“一人一票”的投票结果若用分值来分析如下:
甲(a>c>b)
100 0 0
乙(b>c>a)
100 0 0
丙(c>a>b)
100 0 0
把个人的社会偏好程度完全抹杀掉了,所得的社会偏好次序为:
a=b=c
100 100 100
从而无法整合出群体的意愿排序。
为了回避“一人一票”投票中,候选项目个数达到“三”时因为信息反馈出现盲区而形成的悖论,下面用候选项目“两两对决”的办法进行表决:
若取“a”、“b”对决,那么按照偏好次序排列如下:
甲(a>b)
乙(b>a)
丙(a>b)
社会次序偏好为(a>b)。
若再取“b”、“c”对决,“a”、“c”对决,于是得到三个社会偏 好次序:(a>b)、(b>c)、(c>a),由于这三个关系不符合传递性,所以不构成序关系,因此依然无法获取群体对选项的意愿排序,无法获取群体实现最大满意度的选项。
而“一人一票”或投票人仅仅赋予选项排序所得到的结论:
1)a=b=c
100 100 100
2)(a>b)、(b>c)、(c>a)
都无法正确表达群体意愿。
发明内容
为了解决上述现有技术中的不足,本发明的目的在于提供了一种简便、准确的解读和提取群体意愿的在线统计群体满意度的方法,从而通过网络公平地在线解决争议和进行决策。本发明之所以没有悖论是因为本发明不仅仅要求参与者排序而且要求参与者对选项进行度量化。
为实现上述目的,本发明所采用的技术方案是:
一种在线统计群体满意度的方法,包括以下步骤:
A、确定事件:由统计发起人对事件进行表述,视事件类型,需要指定当事人时,应指定当事人;
B、确定统计群体:统计发起人确定统计在哪个群体进行,使统计发起人在指定的群体中完成满意度统计;
C、确定赋值选项:由群体预先列出各种具体的赋值选项;可由统计发起人自行设定,或者统计发起人向群体征求意见后设定;
D、确定参与赋值者:在群体的全体成员内随机抽取参与赋值者,或者选择群体的全体成员为参与赋值者,确定后通过网络对参与赋值者发送参加赋值的请求;
E、参与赋值者对赋值选项进行满意度赋值:参与赋值者接受赋值请求并进入赋值流程后,对已经确定的赋值选项给出相应的满意度赋值;满意度赋值是参与赋值者的满意度表达,满意度赋值以数值形式或曲线形式存在,满意度值在-N~+N之间,其中N为正实数;
F、根据赋值结果统计出群体最满意的赋值选项:满意度赋值结束后,统计出各赋值选项的群体赋值,对比后得出群体满意度最大的赋值选项,所有赋值选项中满意度最高的赋值选项就是群体最满意的统计结果。
所述步骤C中,赋值选项分为单一性选项和集合性选项;如果某个选项可以利用选项中数值的连续变化来变更选项内容,则称该选项为集合性选项;如果某个选项不能利用选项中数值的连续变化来变更选项内容,则称该选项为单一性选项。根据以上特点,对于单一性选项进行满意度赋值时,只需要赋满意度值即可;对于集合性选项赋值时,则需要确定集合性选项的数值和对应数值的满意度值。赋值选项可以是单一性选项和集合性选项的各种组合。因为人们习惯的赋值选项多种多样,有单一性选项也有集合性选项,但是这样过多的选项不利于比较和实施,为了解决这个问题,需要对各种选项进行等价处理,对各种选项的等价处理属于另一个方法,将在另一个专利申请中提出。
所述步骤E中,N优选10。
所述步骤E中,对于单一性选项,参与赋值者直接对选项给出一个满意度赋值。对于集合性选项,参与赋值者只需对赋值选项的局部进行满意度赋值,然后可以通过设定满意度赋值函数合成出在赋值选项全部定义域里的满意度赋值;从而可以合成出个人满意度赋值的曲线函数。满意度赋值函数必须为单峰有界函数;代表应用为非对称正态分布密度函数。参与赋值者在赋值时,使用三点赋值法或多点赋值法进行赋值,无论使用哪种赋值法,都必须包含最满意选点及该点左右至少一个的赋值点。然后根据赋值者的赋值,利用待定系数法求解出其相应的单峰赋值曲线;也即得到参与赋值者在集合性选项全部定义域里的满意度赋值。若参与赋值者对某一选项不赋值,则把群体赋值作为他的个人赋值。使用三点赋值法时的具体算法可以使用“用非对称正态分布密度函数模拟每个参与赋值者的满意度曲线的算法”,如下:
在以X轴代表赋值选项的数值,Y轴代表满意度的坐标系中,使用两个正态分布密度函数来模拟人的满意度赋值曲线,分别代表满意度上升段和满意度下降段,函数形式为:
Figure PCTCN2014091988-appb-000001
Figure PCTCN2014091988-appb-000002
其中x为赋值选项的数值,f(x)为满意度值,
x∈(-∞,+∞),f(x)∈(-N,+N]
参与赋值者通过三点赋值法确定满意度曲线上的三点。如果最满 意的选项数值为m,选项数值为0的时候满意度为k,最大选项数值max时满意度为v。
则f(m)=N,f(0)=k,f(max)=v
从而可得
a=m
Figure PCTCN2014091988-appb-000003
Figure PCTCN2014091988-appb-000004
至此求出个人的满意度赋值曲线,从而可以确定任意选项数值上的个人满意度。
使用多点赋值法时的具体算法可以使用“用分段函数模拟每个参与赋值者的满意度曲线的算法”,如下:
在以X轴代表赋值选项的数值,Y轴代表满意度的坐标系中,设赋值者在k个点赋值,那就是k+1个分段函数合成出满意度函数。若赋值者在选项点x0、x1、…xh、…、xm处分别给予的满意度为
y0、y1、…yh、…、ym,且x0<x1<…<xh<…<xm,y0<y1<…<yh>…>ym,即赋值者在xh值给予最高满意度数值yh,则该赋值者的满意度函数为:
Figure PCTCN2014091988-appb-000005
针对区间[x0,xm],用“三次样条插值函数”来近似其中的每一段个人满意度函数,“三次样条插值”是数值计算中的常用方法,使用此方法可求出[x0,xm]中每一段函数,此处不再赘述。
针对区间(-∞,x0]与[xm,+∞),用“双曲线函数”
Figure PCTCN2014091988-appb-000006
来近似,分别要求f0(x0)=y0,f0(x0)的一阶导数等于f1(x0)的一阶导数以及fm+1(xm)=ym,fm+1(xm)的一阶导数等于fm(xm)的一阶导数。用待定系数法来确定函数的系数。求出系数后,即可得到所有分段的个人满意度函数f(x),至此求出个人的满意度赋值曲线,从而可以确定任意选项数值上的个人满意度。上述的“样条插值函数”与“双曲线函数”拟合方法是本专利拟合个人低阶光滑满意度函数的可选方法之一。对于要求高阶光滑条件的满意度函数,可以使用更加精细的数学方法来实现,在此不加赘述。
所述步骤F中,对于单一性选项,将参与赋值者对赋值选项的满意度赋值取均值后作为单一性选项的群体赋值;假设有2个个人满意度赋值,y1,y2,那么它们合成的群体赋值就是
Figure PCTCN2014091988-appb-000007
如果有n个个人满意度赋值,y1,y2,…yn,那么它们合成的群体赋值就是
Figure PCTCN2014091988-appb-000008
对于集合性选项,使用计算获得的每个参与赋值者的满意度曲线,合成计算出群体满意度曲线,群体满意度曲线的最高点为集合性选项群体赋值;假设有2个个人满意度曲线,f1(x),f2(x),那么它们合成的群体曲线就是
Figure PCTCN2014091988-appb-000009
如果有n个个人满意度曲线,f1(x),f2(x),…fn(x),那么它们合成的群体满意度曲线就是
Figure PCTCN2014091988-appb-000010
得到最高群体满意度赋值的选项称为该次统计的选中选项。
一种在线统计群体满意度的方法在在线争议解决中的应用,其特征在于,包括以下步骤:
a.确定需要裁决的争议事件:由统计发起人对争议事件进行表述,指定当事人;
b.确定统计群体:统计发起人确定统计在哪个群体进行,使统计发起人在指定的群体中完成统计;由群体对事件或当事人进行判定。
c.确定对当事人可执行的奖惩措施作为赋值选项:由群体预先列出对当事人可执行的奖惩措施作为赋值选项;可由统计发起人自行设定,或者统计发起人向群体征求意见后设定;
d.确定参与赋值者:在群体的全体成员内随机抽取参与赋值者,或者选择群体的全体成员为参与赋值者,确定后通过网络对参与赋值者发送参加赋值的请求;
e.参与赋值者对赋值选项进行满意度赋值:参与赋值者接受赋值请求并进入赋值流程后,对已经确定的赋值选项给出相应的满意度赋值;满意度值在-N~+N之间,其中N为正实数;
f.根据赋值结果统计出群体最满意的赋值选项:满意度赋值结束后,统计出赋值选项的群体赋值,对比后得出群体满意度最大的选项。
所述步骤d中,参与赋值者都是随机抽取的,随机抽样的好处是可以避免网络水军和粉丝团恶意赋值扰乱群体统计结果。
统计数学理论表明,随机抽样统计结果可以代表全体。
准确性的定义:如果能够对该群体的所有成员都发放问卷并且全部合格回收,将其意见整合后便可以得到群体对具体纠纷中不同赋值选项对应不同群体满意度的一个曲线A。如果规定群体最不满意选项对应的满意度为0,群体最满意的选项对应100的话。那么对由某随机抽样产生的满意度曲线而言,如果该抽样统计的最高满意度选项在曲线A中是对应90的话,则称该抽样统计的准确度为90分。
可靠性的定义:如果选取某个抽样方式,其抽样统计结果有99%概率使抽样统计的结论都达到90分以上的话。则称该抽样方式具有99%的可靠性实现90分以上的准确性。
本发明经过实验证明(实验数据详见表1:准确性、可靠性试验),在任意大群体中,当随机抽样人数达到50人,准确性96.08%的可靠性是99%,期望值为99.6%;当随机抽样人数达到80人,准确性98.81%的可靠性是99%,期望值为99.82%;当随机抽样人数达到100人,准确性99.3%的可靠性是99%,期望值为99.88%。
表1:准确性、可靠性试验
总样本数207,每个统计人数从中随机抽取30000次,将随机抽取群体的满意度统计曲线的准确性从高到低排列,表1的具体内容如下:
Figure PCTCN2014091988-appb-000011
因此,本发明中群体对于纠纷处理的决定具有可重复性,该统计的结论具有唯一性。
一种在线统计群体满意度的方法在群体决策中的应用,其特征在于,包括以下步骤:
(a)确定需要决策的事件:由决策发起人作为统计发起人对事件进行表述,视事件类型,需要指定当事人时,应指定当事人;
(b)确定统计群体:统计发起人确定统计在哪个群体进行,使统计发起人在指定的群体中完成统计;
(c)确定需要决策的候选方案作为赋值选项:由群体预先列出各种具体的候选方案作为赋值选项;可由统计发起人自行设定,或者统计发起人向群体征求意见后设定;
(d)确定参与赋值者:在群体的全体成员内随机抽取参与赋值者,或者选择群体的全体成员为参与赋值者,确定后通过网络对参与赋值者发送参加赋值的请求;
(e)参与赋值者对赋值选项进行满意度赋值:参与赋值者接受 赋值请求并进入赋值流程后,对已经确定的赋值选项给出相应的满意度赋值;满意度赋值是参与赋值者的满意度表达,满意度赋值以数值形式或曲线形式存在,满意度值在-N~+N之间,其中N为实数;
(f)根据赋值结果统计出群体最满意的赋值选项:满意度赋值结束后,统计出各赋值选项的群体赋值,对比后得出群体满意度最大的赋值选项即为群体决策的结果。
本发明的有益效果:由于采用上述在线统计群体满意度方法,具有易操作、统计精确的特点,解决了传统方法不能真正表达群体意愿的缺陷;而且由于本发明无需象传统纠纷解决模式那样,引入一个第三方操作者对在线纠纷各方进行调解或仲裁,极大的降低了纠纷解决的成本,提高了纠纷解决的效率。本发明的应用十分广泛,不仅可以应用提案调查、民意调查、群体决策,而且还可以应用于小区管理、网购纠纷、论坛纠纷、Q群纠纷等纠纷处理事件。
附图说明
下面结合附图和具体实施方式对本发明作进一步详细说明:
图1为本发明一种在线统计群体满意度的方法的流程图。
具体实施方式
实施例1:
一种在线统计群体满意度的方法在在线解决网购纠纷处理中的应用。
一种在线统计群体满意度的方法在在线解决争议中的应用,包括以下步骤:
a.确定需要裁决的争议事件:在本纠纠中,购买者某甲和销售者某乙在网络交易中发生了争执,某甲发起解决纠纷请求,描述事件经过,确定事件当事人为某乙,要求对某乙进行裁决,某乙接到通知后,对某甲描述的事件进行补充表述;
b.确定统计群体:统计发起人确定进行统计的群体为交易网站上的所有成员,包括买家、卖家;
c.确定对当事人可执行的奖惩措施作为赋值选项:赋值选项可以有多种,例如赏罚某一数值人民币的金钱,禁止交易某一数值的天数;网站约定以赏罚某一数值人民币的金钱作为赋值选项的代表;
d.确定参与赋值者:在群体的全体成员内随机抽取参与赋值者,利用随机抽样程序,抽取60人参与统计,确定后通过网络对参与赋值者发送参加赋值的请求;
e.参与赋值者对赋值选项进行满意度赋值:满意度范围为[-10,+10]的实数,参与赋值者某丙对罚款人民币2000元的满意度最高为10,罚款人民币1000元的满意度为5,罚款3000元的满意度为6,从而得到3个点;创建满意度坐标系,坐标系中的X轴代表的是赏罚值(单位:元),Y轴代表的是个人对x赏罚值的满意度,根据参与赋值者的赋值结果,使用非对称正态分布密度函数合成参与赋值者在所有定义域上的满意度赋值,并得到个人对惩罚金钱的满意度赋值曲线;
f.根据赋值结果统计出群体最满意的赋值选项:根据获得的每个参与赋值者对赏罚金钱的满意度赋值曲线,合成出群体对赏罚金钱的满意度赋值曲线,此实施例中n为60,在赏罚的满意度曲线中,在罚款1800元选项上群体满意度曲线实现最高满意度数值。满意度最高的点即罚款1800元为群体最满意的赋值选项,也是该在线争议的群体裁决结果。
以上所述是本发明的优选实施方式而已,当然不能以此来限定本发明之权利范围。应当指出,对于本技术领域的普通技术人员来说,对本发明的技术方案进行修改或者等同替换,都不脱离本发明的保护范围。

Claims (10)

  1. 一种在线统计群体满意度的方法,其特征在于,包括以下步骤:A、确定事件:由统计发起人对事件进行表述,视事件类型,需要指定当事人时,应指定当事人;B、确定统计群体:统计发起人确定统计在哪个群体进行,使统计发起人在指定的群体中完成统计;C、确定赋值选项:由群体预先列出各种具体的赋值选项;可由统计发起人自行设定,或者统计发起人向群体征求意见后设定;D、确定参与赋值者:在群体的全体成员内随机抽取参与赋值者,或者选择群体的全体成员为参与赋值者,确定后通过网络对参与赋值者发送参加赋值的请求;E、参与赋值者对赋值选项进行满意度赋值:参与赋值者接受赋值请求并进入赋值流程后,对已经确定的赋值选项给出相应的满意度赋值;满意度赋值是参与赋值者的意愿表达,满意度赋值以数值形式或曲线形式存在,满意度值在-N~+N之间,其中N为正实数;F、根据赋值结果统计出群体最满意的赋值选项:满意度赋值结束后,统计出各赋值选项的群体赋值,对比后得出群体满意度最大的赋值选项,所有赋值选项中满意度最高的赋值选项就是群体最满意的统计结果。
  2. 根据权利要求1所述的一种在线统计群体满意度的方法,其特征在于:所述步骤E中,通过参与赋值者对部分赋值选项进行满意度赋值,以待定系数法合成出参与赋值者在赋值选项全部定义域里的个人的满意度赋值函数,从而合成出个人的满意度赋值曲线。
  3. 根据权利要求2所述的一种在线统计群体满意度的方法,其 特征在于:个人的满意度赋值函数为单峰有界函数。
  4. 根据权利要求3所述的一种在线统计群体满意度的方法,其特征在于:所述赋值选项可以是单一性选项或/和集合性选项。
  5. 根据权利要求4所述的一种在线统计群体满意度的方法,其特征在于:所述步骤E中,使用三点赋值法对集合性选项进行满意度赋值;三点赋值法用于使用非对称正态分布密度函数合成每个参与赋值者的满意度曲线的情况,参与赋值者通过对满意度曲线上的三点进行赋值,来求解函数;这三个点是:取满意度最高的点即满意度为N时的点,满意度最高点两边各取任一个点。
  6. 根据权利要求4所述的一种在线统计群体满意度的方法,其特征在于:所述步骤E中,使用多点赋值法对集合性选项进行满意度赋值;多点赋值法使用连续、光滑、单峰函数合成每个参与赋值者的满意度曲线,赋值者只要给出了满意度最高选点,并且对该点左边任意多点赋值的满意度单增,对该点右边任意多点赋值的满意度单减,则可求解出赋值者的满意度函数。
  7. 根据权利要求1所述的一种在线统计群体满意度的方法,其特征在于:所述步骤E中,N选10。
  8. 根据权利要求5所述的一种在线统计群体满意度的方法,其特征在于:选用非对称正态分布密度函数合成每个参与赋值者的满意度曲线的算法如下:在以X轴代表赋值选项的数值,Y轴代表满意度的坐标系中,使用两个正态分布密度函数来模拟人的满意度赋值曲线,分别代表满意度上升段和满意度下降段,函数形式为:
    Figure PCTCN2014091988-appb-100001
    Figure PCTCN2014091988-appb-100002
    其中x为赋值选项的数值,f(x)为满意度值,x∈(-∞,+∞),f(x)∈(-N,+N];
    参与赋值者通过三点赋值法确定满意度曲线上的三点;这三个点是:取满意度最高的点即满意度为N时的点,满意度最高点两边各取任一个点;将这三个点代入函数,可得三个方程,然后用待定系数法计算出函数的系数,从而得到个人的满意度赋值曲线。
  9. 根据权利要求1所述的一种在线统计群体满意度的方法,其特征在于:所述步骤F中,可以将多个个人满意度赋值或满意度曲线合成出群体的满意度赋值或满意度曲线。
  10. 根据权利要求9所述的一种在线统计群体满意度的方法,其特征在于:所属步骤F中,对于单一性选项,将参与赋值者对赋值选项的满意度赋值取均值后作为单一性选项的群体赋值;对于集合性选项,使用计算获得的每个参与赋值者的满意度曲线,可以合成计算出群体满意度曲线;
    计算单一性选项的群体满意度赋值的算法如下:假设有2个个人满意度赋值,y1,y2,那么它们合成的群体赋值就是
    Figure PCTCN2014091988-appb-100003
    如果有n个个人满意度赋值,y1,y2,…,yn,那么它们合成的群体赋值就是
    Figure PCTCN2014091988-appb-100004
    计算群体的满意度曲线的算法如下:根据获得的多个个体的满意度曲 线,同样在X轴代表的是选项数值,Y轴代表的是个人对x数值的满意度的坐标系中,合成出群体满意度曲线;假设有2个个人满意度曲线,f(x),g(x),那么它们合成的群体曲线就是
    Figure PCTCN2014091988-appb-100005
    如果有n个个人满意度曲线,f1(x),f2(x),…,fn(x),那么它们合成的群体满意度曲线就是
    Figure PCTCN2014091988-appb-100006
PCT/CN2014/091988 2013-11-25 2014-11-24 一种在线统计群体满意度的方法 WO2015074612A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310598815.XA CN103593572B (zh) 2013-11-25 2013-11-25 一种在线统计群体满意度的方法
CN201310598815.X 2013-11-25

Publications (1)

Publication Number Publication Date
WO2015074612A1 true WO2015074612A1 (zh) 2015-05-28

Family

ID=50083709

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/091988 WO2015074612A1 (zh) 2013-11-25 2014-11-24 一种在线统计群体满意度的方法

Country Status (2)

Country Link
CN (1) CN103593572B (zh)
WO (1) WO2015074612A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103593572B (zh) * 2013-11-25 2016-10-05 韩李宾 一种在线统计群体满意度的方法
CN104317268B (zh) * 2014-10-30 2018-03-27 林波荣 一种基于群体满意度定制和节能的建筑室内环境监测、反馈与控制系统及方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070260735A1 (en) * 2006-04-24 2007-11-08 International Business Machines Corporation Methods for linking performance and availability of information technology (IT) resources to customer satisfaction and reducing the number of support center calls
CN102298587A (zh) * 2010-06-24 2011-12-28 深圳市腾讯计算机系统有限公司 满意度调查方法及系统
CN102789496A (zh) * 2012-07-13 2012-11-21 携程计算机技术(上海)有限公司 智能应答的实现方法及系统
CN103593572A (zh) * 2013-11-25 2014-02-19 韩李宾 一种在线统计群体满意度的方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102010010560B3 (de) * 2010-03-05 2011-09-01 Walter Mehnert Verfahren zur Ermittlung des Feinpositionswertes eines bewegbaren Körpers

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070260735A1 (en) * 2006-04-24 2007-11-08 International Business Machines Corporation Methods for linking performance and availability of information technology (IT) resources to customer satisfaction and reducing the number of support center calls
CN102298587A (zh) * 2010-06-24 2011-12-28 深圳市腾讯计算机系统有限公司 满意度调查方法及系统
CN102789496A (zh) * 2012-07-13 2012-11-21 携程计算机技术(上海)有限公司 智能应答的实现方法及系统
CN103593572A (zh) * 2013-11-25 2014-02-19 韩李宾 一种在线统计群体满意度的方法

Also Published As

Publication number Publication date
CN103593572B (zh) 2016-10-05
CN103593572A (zh) 2014-02-19

Similar Documents

Publication Publication Date Title
Huang et al. Aggregation of utility-based individual preferences for group decision-making
Derrick et al. The association between four citation metrics and peer rankings of research influence of Australian researchers in six fields of public health
CN109740036B (zh) Ota平台酒店排序方法及装置
Zhang et al. Determinants of environmental public participation in China: an aggregate level study based on political opportunity theory and post-materialist values theory
Samaha Death and Paperwork Reduction
Fisman et al. The distributional preferences of Americans, 2013–2016
WO2015074612A1 (zh) 一种在线统计群体满意度的方法
Mendoza et al. Quick counts in the Mexican presidential elections: A Bayesian approach
Lamy The shill bidding effect versus the linkage principle
Karam et al. The impact of social media on human resource management scope activities in Al-Futtaim and Al-Etihad group UAE
KR20160138749A (ko) 정책정보 포탈시스템
CN113743796A (zh) 基于权重的多约束条件双随机抽查方法
Luo et al. Parametric prediction from parametric agents
Tideman et al. Paying the partners
Sabir Gender inequality in labour force participation: An empirical investigation
CN111915166A (zh) 一种团体活跃度的确定方法及装置
Zhang et al. DDS: an auction based on a variant of data shapley for federated learning
Cicchi The logic of voting behaviour in the European Parliament: new insights on party group membership and national affiliation as determinants of vote
Sun et al. The structural equation model for public evaluation of the transfer efficiency of rail transit P&R facilities
Abdulhameed Measuring the Arab Parliament’s institutional development
Salois et al. A Generalized B ayesian Instrumental Variable Approach under S tudent t‐distributed Errors with Application
RU2610685C2 (ru) Экспертно-аналитическая информационная система удаленной рейтинговой оценки законопроектов
Schramm et al. A multi-criteria reverse auction to support public purchasing in Brazil
CN110474796B (zh) 基于用户体验质量的异构网络选择方法
Wang Research on risk sharing of PPP project based on shapley value

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: 14863522

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14863522

Country of ref document: EP

Kind code of ref document: A1