CN105184653A - Trust-based crowdsourcing worker screening method for social network - Google Patents
Trust-based crowdsourcing worker screening method for social network Download PDFInfo
- Publication number
- CN105184653A CN105184653A CN201510565022.7A CN201510565022A CN105184653A CN 105184653 A CN105184653 A CN 105184653A CN 201510565022 A CN201510565022 A CN 201510565022A CN 105184653 A CN105184653 A CN 105184653A
- Authority
- CN
- China
- Prior art keywords
- social
- workman
- target
- value
- trust
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a trust-based crowdsourcing worker screening method for a social network. An efficient crowdsourcing worker screening method is provided in a social network environment. The basic framework of a context social network is created. In order to improve the efficiency and the precision of an algorithm, a strong association unit is set, and a novel index structure is constructed for the strong association unit. Modeling is carried out on trusted workers. A trusted crowdsourcing worker selecting algorithm based on contextual information is provided. The method has the advantages of high reliability, less workload, accurate calculation and high reliability.
Description
Technical field
The present invention relates to mass-rent workman and screen field, particularly relate to a kind of mass-rent workman screening technique based on trust towards social networks.
Background technology
Mass-rent is a kind of business model emerging under internet environment.Briefly, mass-rent is that the task that computing machine has been difficult to by mission requirements person is published on network, relies on the process that popular wisdom has been come.But due to the existence of a large amount of tricker's type workman, the confidence level of mass-rent workman is all one of the key issue in mass-rent field all the time.The mass-rent workman of tricker's type to finish the work getable award in order to quick obtaining, often can not pay close attention to task itself, not only cause task not complete by high-quality, mission requirements person also can be made to spend higher time and economic cost.On the contrary, workman trusty makes task to complete by high-quality, for mission requirements person saves time financial cost.Therefore, how in numerous mass-rent workman, selecting trusted workman is the matter of utmost importance solved required for mass-rent field.
The screening technique of existing mass-rent workman mainly lays particular emphasis on the filtration to " tricker's type " workman.Direct scheme uses golden standard data (GoldStandardData).Golden standard data refer to the problem of correct option, can be used for evaluating the trustability of workman.By checking the answer that workman submits to, if workman is by problems erroneous answers, mission requirements person then thinks that this workman is incredible, thus refuses it and finish the work.In addition, some Quality Control Mechanism, by carrying out anticipate to the confidence level of mass-rent workman, according to the confidence criteria that mission requirements person is arranged, thus filter out the substandard workman of confidence level.
At present, the research work chosen about mass-rent workman mainly concentrates on (such as robot of Amazon Turkey (AmazonMechanicalTurk)) on traditional mass-rent platform.Task is distributed on mass-rent platform by mission requirements person, waits for that mass-rent workman has come.A characteristic feature of this kind of mass-rent pattern does not have social networks between mission requirements person and mass-rent workman, causes mission requirements person can only select mass-rent workman passively.But, along with universal (as Facebook, Twitter, the microblogging etc.) of social networks, the content that social networks has that user is numerous, have social networks and share according to user between user, be convenient to infer the features such as its confidence level, mass-rent system combined with social networks and will certainly produce more powerful mass-rent platform.But the screening of existing mass-rent workman mainly through filtering the workman of tricker's type, and does not consider language ambience information.
Summary of the invention
The technical matters that the present invention mainly solves is to provide a kind of mass-rent workman screening technique based on trust towards social networks, have that unfailing performance is high, workload is few, calculating is accurate, confidence level advantages of higher, simultaneously mass-rent workman screening application and universal on have market outlook widely.
For solving the problems of the technologies described above, the technical scheme that the present invention adopts is:
There is provided a kind of mass-rent workman screening technique based on trust towards social networks, step comprises:
(1) build linguistic context social networks framework G=(V, E, LV, LE), wherein, V is vertex set, and E is limit collection, and LV is dependant setpoint attribute, and LE is dependence edge attribute; Described dependant setpoint attribute comprises individual influence, and described dependence edge attribute comprises society and trusts and social cohesion;
(2) in linguistic context social networks, build the subnet of strong associative cell, random choose goes out the summit of the individual social effectiveness containing high numerical value of K in social networks as seed node, for each seed node, depth-first search traversal method is adopted to find the successor node connecting and have high numerical value social influence factors; Each summit of described strong associative cell is connected to the individual influence with higher value, and every bar limit of described strong associative cell is all connected to the social cohesion and society's trust with higher value;
(3) do an index to strong associative cell, described index comprises join index and social context index;
(4) limits value of the trust quality being used as social workman's degree of belief appraisal standard is set, namely the limits value of society's trust, social cohesion and individual influence three social influence factors is set;
(5) arranging source point demander is w1, target workman is expressed as wm, and is expressed as w2 along from source point demander to the intermediate node in the social path of target workman ... wm-1, and society is trusted, social cohesion and this composite value of 3 of individual influence are set to Tp respectively (w1 ..., wm), rp (w1, ..., wm) and ρ p (w1 ..., wm);
(6) composite value and the availability function of 3 social influence factors is calculated:
(6.1) composite value of society's trust: along source point demander to the social trust path of target workman, be multiplied by the social trust value between all intermediate nodes, its computing formula is as follows:
（1）
(6.2) composite value of social cohesion: in a social path, society's cohesion can reduce rapidly along with the increase of intermediate node, in addition, in the social networks of reality, when extending to another summit, society's cohesion can decline faster, and namely weakening of social cohesion is not linear, and its composite value is calculated as follows:
（2）
(6.3) composite value of individual influence: because in social networks, individual influence does not have transitivity, so using the personal influence composite value of the mean value of the personal influence force value of all intermediate nodes as this social path, its account form is as follows:
（3）
(6.4) arrange one as along source point demander to the measurement standard of the degree of belief in the social path of target workman, the availability function Q that the degree of belief being namely used for estimating target workman is measured, its computing formula is as follows:
（4）
Wherein, T, r, ρ are the composite values corresponding to social path effects factor,
,
with
the weight of T, r, ρ respectively,
,
with
equal value in set (0,1) and
+
+
=1, when the availability functional value of a paths is larger, then the confidence level of this target workman is higher;
(7) arrange one to be used for checking whether target workman reaches the objective function δ of source point demander requirement;
If target workman meets the limits value of the trust quality set by source point demander, namely when the social influence factor composite value from source point demander to the social path of target workman is greater than the limits value of corresponding trust quality, objective function computing formula is as follows:
（5）
Wherein, T, r and ρ are the composite value of the social influence factor along source point demander to target workman respectively,
,
,
the limits value of the trust quality set by source point demander respectively;
If trust matter quantitative limitation because a target workman meets, namely
<1, if a target workman does not meet trust matter quantitative limitation, namely
>=1, so target function value is less, the confidence level of target workman is higher;
(8) monte carlo method is adopted, namely by traversing the reverse screening process of source point demander (Vs) from target workman (Vt), and by traversing the selecting due process of target workman (Vt) from source point demander (Vs), find the target workman meeting source point demander and require.
In a preferred embodiment of the present invention, described join index have recorded the information on a series of summit, the index on each summit contains ancestors' node and the successor node on this summit, when the summit that will inquire about is included in strong associative cell, can directly check its join index, find its ancestors' node or successor node.
In a preferred embodiment of the present invention, the step of described social context index is: calculate the social trust value of the mulitpath between two summits, social cohesion and individual influence composite value, if three of a paths social influence factors values, namely social trust value, social cohesion and individual influence are all greater than all the other paths, so social context index will record the social influence factors value synthesized by this paths, otherwise, the path of the maximal value of society's trust, social cohesion and individual influence is had respectively with regard to index three.
In a preferred embodiment of the present invention, described reverse screening, namely from target workman point, checks that can target workman meet the trust quality limitations of source point demander setting, and screening falls can not meet the target workman of source point demander primary demand, and its concrete steps comprise:
A () searches for K strong associative cell in order, if search successfully, namely in strong associative cell, corresponding summit is found, then its ancestors' node is added to and oppositely mark in vertex set, be saved in its ancestor node by from target workman node to the corresponding composite value of ancestors' node on the summit that will search; If search unsuccessfully, carry out step (b);
B () calculates and puts the current target function value choosing the social path of all of its neighbor point a little from target workman, and produce the candidate point of K minimum target functional value;
C () picks out one as the point that will expand in candidate point, and the new probability formula selecting the point of expansion is as follows:
(6);
(d) preserve corresponding T, r,
composite value is to current extensions point;
If e () does not arrive the target workman point that source point demander requires, then continue step (a); If δ >1, then filter out this target workman point; If δ≤1, then added to candidate workman and concentrated.
In a preferred embodiment of the present invention, described selecting due, exactly from source point demander, check it is how reliable that a target workman has actually, and namely check from source point demander and have much to its getable usable levels in social path, its concrete steps comprise:
F () searches for the strong associative cell of K in order, if the summit that will inquire about is included in strong associative cell, then added to by the successor node on summit in forward expansion vertex set, and preserve corresponding T, r,
composite value to the successor node on summit, otherwise, carry out step (g);
(g) calculate from source point demander to current extensions point and be reversed the availability value of labeled abutment points, and pick out the point alternatively point that K has maximum availability value;
H () picks out one as the point that will expand in candidate point, and the new probability formula selecting the point of expansion is as follows:
(7);
(i) preserve corresponding T, r,
composite value is to current extensions point;
If j () does not arrive target workman point, continue step (f), otherwise, return its usable levels.
The invention has the beneficial effects as follows:.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, below the accompanying drawing used required in describing embodiment is briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings, wherein:
Fig. 1 be of the present invention a kind of towards social networks based on trust mass-rent workman screening technique in linguistic context social networks structural representation;
Fig. 2 is the structural representation of strong associative cell in the present invention;
Fig. 3 is the structural representation of strong associative cell index in the present invention.
Embodiment
Be clearly and completely described to the technical scheme in the embodiment of the present invention below, obviously, described embodiment is only a part of embodiment of the present invention, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making other embodiments all obtained under creative work prerequisite, belong to the scope of protection of the invention.
Refer to Fig. 1-3, the embodiment of the present invention comprises:
Towards the mass-rent workman screening technique based on trust of social networks, step comprises:
(1) basic framework is created: build linguistic context social networks framework G=(V, E, LV, LE), wherein, V is vertex set, and E is limit collection, and LV is dependant setpoint attribute, and LE is dependence edge attribute; Described dependant setpoint attribute comprises individual influence, and described dependence edge attribute comprises society and trusts and social cohesion.
Social influence factor comprises individual influence, society trusts and social cohesion.
Social to trust: it is the conviction of a social user for another social user that society trusts, according to the contact between them, and to a certain extent, following action be every-formed by the latter will cause expected result.Obviously, in social networks, the trust value between two social users has marked difference in different field, and such as, user A trusts the English proficiency of user B, but may distrust that B repairs the ability of automobile.Represented in specific area by TAB [0,1], social user A, to the trust value of social user B, if TAB=0, represents that A distrusts B completely; On the contrary, if TAB=1, then represent that A thoroughly believes the ability of B in this field.
Social cohesion: The Study of Sociology finds, a social user more can trust and have more social bonds to have the social user of more close relationship with these.Represented in social networks by rAB [0,1], social user A, to the cohesion of social user B, if rAB=0, represents do not have close relationship between A and B; On the contrary, if rAB=1, then represent that A and B has the most intimate social networks.
Individual influence: in social networks, the social activities that social participant enriches can be divided into different fields (such as, recruitment person or sellers).Sociology shows, in a specific field, the suggestion of an expert is more more reliable than a beginner.Therefore, will
a [0,1] represents the social effectiveness of social user A at a specific area, and individual social effectiveness is determined by the professional degree of user.
a=0 represents that A does not understand this field, does not have influence power completely, on the contrary
a=1 represents that A is an expert in this field, has the highest influence power.
(2) concept of strong associative cell, and be the index structure that strong associative cell constructs a kind of novelty.(2.1) in linguistic context social networks, build the subnet of strong associative cell, random choose goes out the summit of the individual social effectiveness containing high numerical value of K in social networks as seed node, for each seed node, depth-first search traversal method is adopted to find the successor node connecting and have high numerical value social influence factors; Each summit of described strong associative cell is connected to the individual influence with higher value, and every bar limit of described strong associative cell is all connected to the social cohesion and society's trust with higher value.
Strong associative cell StrongSocialComponent): a strong associative cell is a strongly connected subnet in linguistic context social networks, its each summit is connected to larger individual influence, and every bar limit is all connected to larger cohesion and social trusting relationship.
Identify that in a social networks, all strong associative cells can spend larger time cost.Therefore, we optionally identify K strong associative cell wherein, and way is as follows:
First random choose goes out K in social networks the summit containing higher social effectiveness as seed node, for each seed node, adopts depth-first search traversal method to find the successor node being connected to higher social influence factors value.
(2.2) according to the analysis of social science, in a strong associative cell, social structure and social effectiveness factor can keep stable in longer time cycle.This makes us can do an index with lower renewal cost to strong associative cell.
(2.3) do an index to strong associative cell, can improve efficiency and the precision of mass-rent workman Algorithms of Selecting, described index comprises join index and social context index.
Join index: this index record information on a series of summit, the index on each summit contains ancestors' node and the successor node on this summit.Such as, in figure 3, for summit B, because the existing ancestors' node of B has successor node again, therefore, join index have recorded ancestors' node of B, i.e. Anc:A, and the successor node of B, that is, Des:E, equally, our join index that has been other node structures in strong associative cell.If the summit inquired about is included in strong associative cell, we can directly check its join index, find its ancestors' node or successor node fast, can reduce query time in a large number.
Social context index: the maximal value of the social influence factors of this index record synthesis.Specific explanations is as follows: have mulitpath between two summits and pass through, if paths three social influence factors (i.e. society's trust, social cohesion and individual influence) is all greater than all the other paths, so social context index will record the social influence factors value synthesized by this paths.Otherwise, the path that we have society's trust, social cohesion and individual influence composite value maximum with regard to index three respectively.
In figure 3, for summit A, there are two paths processes from summit A to summit E, i.e. path p1 (A, B, E) and p2 (A, C, E).Due to large than path p2 (A, C, E) of the social influence factors value of path p1 (A, B, E), therefore we are in the social influence factors value on summit E record path p1 (A, B, E), i.e. AS={0.96,0.88,0.91}.Equally, we are all the other summits structure social context index.
(3) modeling has been carried out to trusted workman, and proposed the Algorithms of Selecting of the trusted mass-rent workman based on language ambience information.
(3.1) limits value of the trust quality being used as social workman's degree of belief appraisal standard is set, namely the limits value of society's trust, social cohesion and individual influence three social influence factors is set.
Trust quality (QoT): trust quality and refer to consideration society trust, social cohesion and individual influence three social influence factors, along a social trust path, in trust recommendation, reach certain level of trust.
In social workman's Selection Model, a mission requirements person can arrange the standard of different QoT restrictions as social workman's degree of belief appraisal for social workman, such as, in FIG, mission requirements person B can arrange QoT and limit λ={ λ T>0.5, λ r>0.5, λ ρ >0.5}, wherein λ T, λ r, λ ρ is T respectively, the limitation standard of r, ρ. obviously, social workman D can not meet the degree of belief standard of B, therefore, B can not select D to finish the work.
(3.2) arranging source point demander is w1, target workman is expressed as wm, and is expressed as w2 along from source point demander to the intermediate node in the social path of target workman ... wm-1, and society is trusted, social cohesion and this composite value of 3 of individual influence are set to Tp respectively (w1 ..., wm), rp (w1, ..., wm) and ρ p (w1 ..., wm).
(3.3) composite value and the availability function of 3 social influence factors is calculated:
(A) composite value of society's trust: along source point demander to the social trust path of target workman, be multiplied by the social trust value between all intermediate nodes, its computing formula is as follows:
（1）
(B) composite value of social cohesion: in a social path, society's cohesion can reduce rapidly along with the increase of intermediate node, in addition, in the social networks of reality, when extending to another summit, society's cohesion can decline faster, and namely weakening of social cohesion is not linear, and its composite value is calculated as follows:
（2）
(C) composite value of individual influence: because in social networks, individual influence does not have transitivity, so using the personal influence composite value of the mean value of the personal influence force value of all intermediate nodes as this social path, its account form is as follows:
（3）
(D) arrange one as along source point demander to the measurement standard of the degree of belief in the social path of target workman, the availability function Q that the degree of belief being namely used for estimating target workman is measured, its computing formula is as follows:
（4）
Wherein, T, r, ρ are the composite values corresponding to social path effects factor,
,
with
the weight of T, r, ρ respectively,
,
with
equal value in set (0,1) and
+
+
=1, when the availability functional value of a paths is larger, then the confidence level of this target workman is higher.
(3.4) arrange one to be used for checking whether target workman reaches the objective function δ of source point demander requirement;
If target workman meets the limits value of the trust quality set by source point demander, namely when the social influence factor composite value from source point demander to the social path of target workman is greater than the limits value of corresponding trust quality, objective function computing formula is as follows:
（5）
Wherein, T, r and ρ are the composite value of the social influence factor along source point demander to target workman respectively,
,
,
the limits value of the trust quality set by source point demander respectively;
If trust matter quantitative limitation because a target workman meets, namely
<1, if a target workman does not meet trust matter quantitative limitation, namely
>=1, so target function value is less, the confidence level of target workman is higher.
(3.5) monte carlo method is adopted, namely by traversing source point demander (Vs) from target workman (Vt)
Reverse screening process, and by traverse the selecting due process of target workman (Vt) from source point demander (Vs), find the target workman meeting source point demander and require.
Described reverse screening, namely from target workman point, check that can target workman meet the QoT restriction of source point demander setting, and screening falls can not meet the target workman of source point demander primary demand, its concrete steps comprise:
A () searches for the strong associative cell of K in order, if search successfully, added to by its ancestors' node oppositely in mark vertex set, and preserve corresponding T, r and
composite value to its ancestors' node, otherwise, carry out step (b);
B () calculates and puts the current target function value choosing the social path of all of its neighbor point a little from target workman, and produce the candidate point of K minimum target functional value;
C () picks out one as the point that will expand in candidate point, and the new probability formula selecting the point of expansion is as follows:
(6);
(d) preserve corresponding T, r,
composite value is to current extensions point;
If e () does not arrive the target workman point that source point demander requires, then continue step (a); If δ >1, then filter out this target workman point; If δ≤1, then added to candidate workman and concentrated.
Described selecting due, exactly from source point demander, check it is how reliable that a target workman has actually, and namely check from source point demander and have much to its getable usable levels in social path, its concrete steps comprise:
F () searches for the strong associative cell of K in order, if the summit that will inquire about is included in strong associative cell, then added to by the successor node on summit in forward expansion vertex set, and preserve corresponding T, r,
composite value to the successor node on summit, otherwise, carry out step (g);
(g) calculate from source point demander to current extensions point and be reversed the availability value of labeled abutment points, and pick out the point alternatively point that K has maximum availability value;
H () picks out one as the point that will expand in candidate point, and the new probability formula selecting the point of expansion is as follows:
(7);
(i) preserve corresponding T, r,
composite value is to current extensions point;
If j () does not arrive target workman point, continue step (f), otherwise, return the value of its availability function.
A kind of beneficial effect based on the mass-rent workman screening technique of trusting towards social networks of the present invention is: the language ambience information hidden based on social networks, pick out social workman trusty, efficiency is high, effectively reduce workload, increase work efficiency, and calculate accurate, with a high credibility, the screening technique of different demand can be met.
The foregoing is only embodiments of the invention; not thereby the scope of the claims of the present invention is limited; every utilize description of the present invention to do equivalent structure or equivalent flow process conversion; or be directly or indirectly used in other relevant technical field, be all in like manner included in scope of patent protection of the present invention.
Claims (5)
1., towards the mass-rent workman screening technique based on trust of social networks, it is characterized in that, step comprises:
(1) build linguistic context social networks framework G=(V, E, LV, LE), wherein, V is vertex set, and E is limit collection, and LV is dependant setpoint attribute, and LE is dependence edge attribute; Described dependant setpoint attribute comprises individual influence, and described dependence edge attribute comprises society and trusts and social cohesion;
(2) in linguistic context social networks, build the subnet of strong associative cell, the integer K that random choose goes out in social networks contains the summit of the social effectiveness of high numerical value as seed node, for each seed node, depth-first search traversal method is adopted to find the successor node connecting and have high numerical value social influence factors; Each summit of described strong associative cell is connected to the individual influence with higher value, and every bar limit of described strong associative cell is all connected to the social cohesion and society's trust with higher value;
(3) do an index to strong associative cell, described index comprises join index and social context index;
(4) limits value of the trust quality being used as social workman's degree of belief appraisal standard is set, namely the limits value of society's trust, social cohesion and individual influence three social influence factors is set;
(5) arranging source point demander is w1, target workman is expressed as wm, and is expressed as w2 along from source point demander to the intermediate node in the social path of target workman ... wm-1, and society is trusted, social cohesion and this composite value of 3 of individual influence are set to Tp respectively (w1 ..., wm), rp (w1, ..., wm) and ρ p (w1 ..., wm);
(6) composite value and the availability function of 3 social influence factors is calculated:
(6.1) composite value of society's trust: along source point demander to the social trust path of target workman, be multiplied by the social trust value between all intermediate nodes, its computing formula is as follows:
（1）
(6.2) composite value of social cohesion: in a social path, society's cohesion can reduce rapidly along with the increase of intermediate node, in addition, in the social networks of reality, when extending to another summit, society's cohesion can decline faster, and namely weakening of social cohesion is not linear, and its composite value is calculated as follows:
（2）
(6.3) composite value of individual influence: because in social networks, individual influence does not have transitivity, so using the personal influence composite value of the mean value of the personal influence force value of all intermediate nodes as this social path, its account form is as follows:
（3）
(6.4) arrange one as along source point demander to the measurement standard of the degree of belief in the social path of target workman, the availability function Q that the degree of belief being namely used for estimating target workman is measured, its computing formula is as follows:
（4）
Wherein, T, r, ρ are the composite values corresponding to social path effects factor,
,
with
the weight of T, r, ρ respectively,
,
with
equal value in set (0,1) and
+
+
=1, when the availability functional value of a paths is larger, then the confidence level of this target workman is higher;
(7) arrange one to be used for checking whether target workman reaches the objective function δ of source point demander requirement;
If target workman meets the limits value of the trust quality set by source point demander, namely when the social influence factor composite value from source point demander to the social path of target workman is greater than the limits value of corresponding trust quality, objective function computing formula is as follows:
（5）
Wherein, T, r and ρ are the composite value of the social influence factor along source point demander to target workman respectively,
,
,
the limits value of the trust quality set by source point demander respectively;
If trust matter quantitative limitation because a target workman meets, namely
<1, if a target workman does not meet trust matter quantitative limitation, namely
>=1, so target function value is less, the confidence level of target workman is higher;
(8) monte carlo method is adopted, namely by traversing the reverse screening process of source point demander (Vs) from target workman (Vt), and by traversing the selecting due process of target workman (Vt) from source point demander (Vs), find the target workman meeting source point demander and require.
2. a kind of mass-rent workman screening technique based on trust towards social networks according to claim 1, it is characterized in that, described join index have recorded the information on a series of summit, the index on each summit contains ancestors' node and the successor node on this summit, when the summit that will inquire about is included in strong associative cell, can directly check its join index, find its ancestors' node or successor node.
3. a kind of mass-rent workman screening technique based on trust towards social networks according to claim 1, it is characterized in that, the step of described social context index is: the social trust value calculating the mulitpath between two summits, society's cohesion and individual influence composite value, if three of a paths social influence factors values, i.e. social trust value, society's cohesion and individual influence are all greater than all the other paths, so social context index will record the social influence factors value synthesized by this paths, otherwise, society is had to trust respectively with regard to index three, the path of the maximal value of society's cohesion and individual influence.
4. a kind of mass-rent workman screening technique based on trust towards social networks according to claim 1, it is characterized in that, described reverse screening, namely from target workman point, check that can target workman meet the trust quality limitations of source point demander setting, and screening falls can not meet the target workman of source point demander primary demand, and its concrete steps comprise:
A () searches for K strong associative cell in order, if search successfully, namely in strong associative cell, corresponding summit is found, then its ancestors' node is added to and oppositely mark in vertex set, be saved in its ancestor node by from target workman node to the corresponding composite value of ancestors' node on the summit that will search; If search unsuccessfully, carry out step (b);
B () calculates and puts the current target function value choosing the social path of all of its neighbor point a little from target workman, and produce the candidate point of K minimum target functional value;
C () picks out one as the point that will expand in candidate point, and the new probability formula selecting the point of expansion is as follows:
(6);
(d) preserve corresponding T, r,
composite value is to current extensions point;
If e () does not arrive the target workman point that source point demander requires, then continue step (a); If δ >1, then filter out this target workman point; If δ≤1, then added to candidate workman and concentrated.
5. a kind of mass-rent workman screening technique based on trust towards social networks according to claim 1, it is characterized in that, described selecting due, exactly from source point demander, check that how reliable a target workman have actually, namely checking from source point demander has much to its getable usable levels in social path, and its concrete steps comprise:
F () searches for the strong associative cell of K in order, if the summit that will inquire about is included in strong associative cell, then added to by the successor node on summit in forward expansion vertex set, and preserve corresponding T, r,
composite value to the successor node on summit, otherwise, carry out step (g);
(g) calculate from source point demander to current extensions point and be reversed the availability value of labeled abutment points, and pick out the point alternatively point that K has maximum availability value;
H () picks out one as the point that will expand in candidate point, and the new probability formula selecting the point of expansion is as follows:
(7);
(i) preserve corresponding T, r,
composite value is to current extensions point;
If j () does not arrive target workman point, continue step (f), otherwise, return its usable levels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510565022.7A CN105184653A (en) | 2015-09-08 | 2015-09-08 | Trust-based crowdsourcing worker screening method for social network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510565022.7A CN105184653A (en) | 2015-09-08 | 2015-09-08 | Trust-based crowdsourcing worker screening method for social network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105184653A true CN105184653A (en) | 2015-12-23 |
Family
ID=54906710
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510565022.7A Pending CN105184653A (en) | 2015-09-08 | 2015-09-08 | Trust-based crowdsourcing worker screening method for social network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105184653A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106228029A (en) * | 2016-08-30 | 2016-12-14 | 北京航空航天大学 | Quantification problem method for solving based on mass-rent and device |
| CN107871196A (en) * | 2016-09-28 | 2018-04-03 | 郑州大学 | A kind of mass-rent method for evaluating quality based on slip task window |
| CN109460490A (en) * | 2018-09-28 | 2019-03-12 | 上海达家迎信息科技有限公司 | Determination method, apparatus, equipment and the storage medium of culminating point in social networks |
| CN110263372A (en) * | 2019-05-20 | 2019-09-20 | 上海大学 | A kind of event detecting method based on intelligent perception system |
| WO2019200736A1 (en) * | 2018-04-17 | 2019-10-24 | 平安科技（深圳）有限公司 | Operating method and device for crowdsourcing platform, computer device and storage medium |
| CN111292062A (en) * | 2020-02-10 | 2020-06-16 | 中南大学 | Crowdsourcing garbage worker detection method and system based on network embedding and storage medium |
| CN111444332A (en) * | 2020-03-13 | 2020-07-24 | 广州大学 | Crowdsourcing worker reliability model establishing method and device under crowdsourcing knowledge verification environment |
| CN113032426A (en) * | 2021-04-08 | 2021-06-25 | 平安科技（深圳）有限公司 | Intelligent verification method, device and equipment for identification result and storage medium |
| WO2023010880A1 (en) * | 2021-08-03 | 2023-02-09 | 华为技术有限公司 | Data transmission method and related device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102456062A (en) * | 2010-11-04 | 2012-05-16 | 中国人民解放军国防科学技术大学 | Community similarity calculation method and social network cooperation mode discovery method |
| CN102546754A (en) * | 2011-11-23 | 2012-07-04 | 河南理工大学 | Service quality customized Web service combination method |
| CN103268520A (en) * | 2013-05-09 | 2013-08-28 | 武汉大学 | Network team automatic building method based on skill contribution values |
-
2015
- 2015-09-08 CN CN201510565022.7A patent/CN105184653A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102456062A (en) * | 2010-11-04 | 2012-05-16 | 中国人民解放军国防科学技术大学 | Community similarity calculation method and social network cooperation mode discovery method |
| CN102546754A (en) * | 2011-11-23 | 2012-07-04 | 河南理工大学 | Service quality customized Web service combination method |
| CN103268520A (en) * | 2013-05-09 | 2013-08-28 | 武汉大学 | Network team automatic building method based on skill contribution values |
Non-Patent Citations (2)
| Title |
|---|
| GUANFENG LIU 等: ""Discovering Trust Networks for the Selection of Trustworthy Service Providers in Complex Contextual Social Networks"", 《WEB SERVICES (ICWS), 2012 IEEE 19TH INTERNATIONAL CONFERENCE ON》 * |
| GUANFENG LIU 等: "《DATABASE SYSTEMS FOR ADVANCED APPLICATIONS, DASFAA 2015, PT II》", 9 April 2015 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106228029B (en) * | 2016-08-30 | 2019-02-22 | 北京航空航天大学 | Quantification problem method for solving and device based on crowdsourcing |
| CN106228029A (en) * | 2016-08-30 | 2016-12-14 | 北京航空航天大学 | Quantification problem method for solving based on mass-rent and device |
| CN107871196A (en) * | 2016-09-28 | 2018-04-03 | 郑州大学 | A kind of mass-rent method for evaluating quality based on slip task window |
| WO2019200736A1 (en) * | 2018-04-17 | 2019-10-24 | 平安科技（深圳）有限公司 | Operating method and device for crowdsourcing platform, computer device and storage medium |
| CN109460490B (en) * | 2018-09-28 | 2021-05-04 | 上海达家迎信息科技有限公司 | Method, device, equipment and storage medium for determining center vertex in social network |
| CN109460490A (en) * | 2018-09-28 | 2019-03-12 | 上海达家迎信息科技有限公司 | Determination method, apparatus, equipment and the storage medium of culminating point in social networks |
| CN110263372A (en) * | 2019-05-20 | 2019-09-20 | 上海大学 | A kind of event detecting method based on intelligent perception system |
| CN110263372B (en) * | 2019-05-20 | 2023-05-23 | 上海大学 | Event detection method based on crowd sensing system |
| CN111292062A (en) * | 2020-02-10 | 2020-06-16 | 中南大学 | Crowdsourcing garbage worker detection method and system based on network embedding and storage medium |
| CN111444332B (en) * | 2020-03-13 | 2023-04-18 | 广州大学 | Crowdsourcing worker reliability model establishing method and device under crowdsourcing knowledge verification environment |
| CN111444332A (en) * | 2020-03-13 | 2020-07-24 | 广州大学 | Crowdsourcing worker reliability model establishing method and device under crowdsourcing knowledge verification environment |
| CN113032426A (en) * | 2021-04-08 | 2021-06-25 | 平安科技（深圳）有限公司 | Intelligent verification method, device and equipment for identification result and storage medium |
| WO2023010880A1 (en) * | 2021-08-03 | 2023-02-09 | 华为技术有限公司 | Data transmission method and related device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105184653A (en) | Trust-based crowdsourcing worker screening method for social network | |
| CN101388024B (en) | Compression space high-efficiency search method based on complex network | |
| Wang | A fuzzy multi-criteria decision-making model by associating technique for order preference by similarity to ideal solution with relative preference relation | |
| Mardani et al. | Determining the utility in management by using multi-criteria decision support tools: a review | |
| CN110213164B (en) | Method and device for identifying network key propagator based on topology information fusion | |
| CN102915510B (en) | Power Project network post evaluation system based on multi-hierarchical fuzzy comprehensive evaluation | |
| CN102880657A (en) | Expert recommending method based on searcher | |
| Zhang et al. | Mechanism design for finding experts using locally constructed social referral web | |
| Wang et al. | A new method for two-sided matching decision making of PPP projects based on intuitionistic fuzzy choquet integral | |
| KR20180089479A (en) | User data sharing method and device | |
| Deng et al. | A novel mean-variance-maverick DEA prospect cross-efficiency approach for fuzzy portfolio selection | |
| Zhang et al. | A reinforcement learning-based stakeholder value aggregation model for collaborative decision making on disaster resilience | |
| Wang et al. | Cross-efficiency intervals integrated ranking approach based on the generalized Fermat-Torricelli point | |
| CN105786622A (en) | Node selection method for real-time cooperative computing in cloud environment | |
| Ren et al. | Switched Surplus-Based Distributed Security Dispatch for Smart Grid With Persistent Packet Loss | |
| Zhou et al. | Black-box attacks against signed graph analysis via balance poisoning | |
| Barhum | Tight bounds for the advice complexity of the online minimum steiner tree problem | |
| CN116070890A (en) | Workflow optimization method based on virtualized production quality under time constraint | |
| CN112966155B (en) | Link prediction method based on path correlation | |
| CN107018027A (en) | Link prediction method based on Bayesian estimation and common neighbor node degree | |
| Chen et al. | Introduction: ASEAN development in the digital economy | |
| CN112288007A (en) | Cross-domain identity association method and system based on multivariate relational expression | |
| Tiwari et al. | Smart contract for decentralized water management system using blockchain technology | |
| Xu et al. | An efficient trust propagation scheme for predicting trustworthiness of service providers in service-oriented social networks | |
| ELMA | RENEWABLE ENERGY COMPANIES’RECENT OUTLOOK ON BORSA ISTANBUL WITH MOORA AND MABAC METHODS |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151223 |
|
| RJ01 | Rejection of invention patent application after publication |