Method and system for providing weighted evaluation
Systems and methods are disclosed for providing weighted evaluation. The system may comprise one or more processors and a memory storing instructions that, when executed by the one or more processors, cause the system to obtain a plurality of data point groups, obtain an inbound rank of each data point group based on one or more inbound ranks of one or more data groups linking to the data point group, obtain an outbound rank of the each data point group based on one or more outbound ranks of one or more data groups linked from the each data point group, obtain a group rank of the each data point group based on the inbound rank and the outbound rank of the each data point group, and process the data point groups according to the corresponding group ranks.
1. A system for providing weighted evaluation, the system comprising:
one or more processors; and
a memory storing instructions that, when executed by the one or more processors, cause the system to:
obtain a plurality of data point groups, wherein each data point group comprises one or more data points, one or more of the data points are linked data points, one or more of the data points are linking data points, and each of the linked data points is pointed to by a link from one or more of the linking data points;
obtain an inbound rank of each data point group based on one or more inbound ranks of one or more data groups linking to the data point group;
obtain an outbound rank of the each data point group based on one or more outbound ranks of one or more data groups linked from the each data point group;
obtain a group rank of the each data point group based on the inbound rank and the outbound rank of the each data point group, wherein the inbound rank and the outbound rank of the each data point group are weighted based on an inbound-outbound balance coefficient; and
process the data point groups according to the corresponding group ranks.
2. The system of claim 1 , wherein:
to obtain the inbound rank of the data point group, the system is caused to iteratively obtain the inbound rank based on one or more inbound ranks of one or more other data point groups until a change in the iteration is below a preset threshold; and
to obtain the outbound rank of the data point group, the system is caused to iteratively obtain the outbound rank based on one or more outbound ranks of one or more other data point groups until a change in the iteration is below another preset threshold.
3. The system of claim 1 , wherein:
the plurality of data point groups comprise a first and a second data point groups; and
to process the first and the second data point groups, the system is further caused to:
determine a connection weight between the first data point group and the second data point group; and
in response to determining the connection weight being over a preset threshold, merge the first and the second data point groups to obtain a merged data point group.
4. The system of claim 3 , wherein:
the connection weight between the first data point group A and the second data point group B comprises W(AB) and W(BA);
W(AB) is based on an interaction weight I(AB) and a quality weight Q(AB);
the interaction weight I(AB) represents a closeness of connection between one or more data points in A and one or more data points in B; and
the quality weight Q(AB) represents a degree of evaluation of one or more data points in B by one or more data points in A.
5. The system of claim 3 , wherein:
the connection weight between the first and the second data point groups is larger than any other connection weight between the first data point group and another data point group.
6. The system of claim 3 , wherein:
to process the first and the second data point groups, the system is further caused to:
obtain an inbound rank and an outbound rank of the merged data point group;
update inbound and outbound ranks of all of the data point groups other than the first and second point groups; and
remove the first and second data point groups.
7. The system of claim 6 , wherein:
to process the first and the second data point groups, the system is further caused to obtain a connection weight of the merged data point group; and
the connection weight of the merged data point group is a weighted average of a connection weight with respect to the first data point group and another connection weight with respect to the second data point group.
8. The system of claim 7 , wherein:
the preset threshold for the connection weight is based on a computing method of the connection weight of the merged data point group.
9. The system of claim 6 , wherein:
the system is further caused to perform the connection weight determination and the merge for two or more other data point groups.
10. A computer-implemented method, the method being implemented by a computing system including one or more processors and storage media storing machine-readable instructions, the method comprising:
obtaining a plurality of data point groups, wherein each data point group comprises one or more data points, one or more of the data points are linked data points, one or more of the data points are linking data points, and each of the linked data points is pointed to by a link from one or more of the linking data points;
obtaining an inbound rank of each data point group based on one or more inbound ranks of one or more data groups linking to the data point group;
obtaining an outbound rank of the each data point group based on one or more outbound ranks of one or more data groups linked from the each data point group;
obtaining a group rank of the each data point group based on the inbound rank and the outbound rank of the each data point group, wherein the inbound rank and the outbound rank of the each data point group are weighted based on an inbound-outbound balance coefficient; and
processing the data point groups according to the corresponding group ranks.
11. The computer-implemented method of claim 10 , wherein:
obtaining the inbound rank of the data point group comprises iteratively obtaining the inbound rank based on one or more inbound ranks of one or more other data point groups until a change in the iteration is below a preset threshold; and
obtaining the outbound rank of the data point group comprises iteratively obtaining the outbound rank based on one or more outbound ranks of one or more other data point groups until a change in the iteration is below another preset threshold.
12. The computer-implemented method of claim 10 , wherein:
the plurality of data point groups comprise a first and a second data point groups; and
processing the first and the second data point groups comprises:
determining a connection weight between the first data point group and the second data point group; and
in response to determining the connection weight being over a preset threshold, merging the first and the second data point groups to obtain a merged data point group.
13. The computer-implemented method of claim 12 , wherein:
the connection weight between the first data point group A and the second data point group B comprises W(AB) and W(BA);
W(AB) is based on an interaction weight I(AB) and a quality weight Q(AB);
the interaction weight I(AB) represents a closeness of connection between one or more data points in A and one or more data points in B; and
the quality weight Q(AB) represents a degree of evaluation of one or more data points in B by one or more data points in A.
14. The computer-implemented method of claim 12 , wherein:
the connection weight between the first and the second data point groups is larger than any other connection weight between the first data point group and another data point group.
15. The computer-implemented method of claim 12 , wherein:
processing the first and the second data point groups comprises:
obtain an inbound rank and an outbound rank of the merged data point group;
update inbound and outbound ranks of all of the data point groups other than the first and second point groups; and
remove the first and second data point groups.
16. The computer-implemented method of claim 15 , wherein:
processing the first and the second data point groups comprises obtaining a connection weight of the merged data point group; and
the connection weight of the merged data point group is a weighted average of a connection weight with respect to the first data point group and another connection weight with respect to the second data point group.
17. The computer-implemented method of claim 16 , wherein:
the preset threshold for the connection weight is based on a computing method of the connection weight of the merged data point group.
18. The computer-implemented method of claim 15 , further comprising:
performing the connection weight determination and the merge for two or more other data point groups.
19. A non-transitory computer readable medium comprising instructions that, when executed, cause one or more processors to:
obtain a plurality of data point groups comprising a first data point group and a second data point group, wherein each data point group comprises one or more data points, one or more of the data points are linked data points, one or more of the data points are linking data points, each of the linked points is pointed to by a link from one or more of the linking data points, each data point group has an inbound rank and an outbound rank, and each data point group has a group rank based on the inbound rank and the outbound rank weighted based on an inbound-outbound balance coefficient;
determine a connection weight between the first data point group and the second data point group, wherein the connection weight is based on a closeness of connection between one or more data points in the first data point group and one or more data points in the second data point group and based on a degree of evaluation of one or more data points from the first and second data point groups; and
in response to determining the connection weight being over a preset threshold, merge the first and the second data point groups to obtain a merged data point group.
20. The non-transitory computer readable medium of claim 19 , wherein:
the first data point group is A and the second data point group is B,
the connection weight is W(AB)+W(BA);
W(AB) is based on an interaction weight I(AB) and a quality weight Q(AB);
W(BA) is based on an interaction weight I(BA) and a quality weight Q(BA);
the interaction weight I(AB) and I(BA) each represents a closeness of connection between one or more data points in A and one or more data points in B; and
the quality weight Q(AB) represents a degree of evaluation of one or more data points in B by one or more data points in A; and
the quality weight Q(BA) represents a degree of evaluation of one or more data points in A by one or more data points in B.