System and method for presentation and visual navigation of network data sets

1. A computer implemented method of determining a similarity score between two data records in a data record set, each data record including a predetermined data record strength value representative of the importance of each respective data record in the data record set, the method comprising:

identifying a target data record from the data record set;

identifying data records within the data set having primary and secondary connections with the target record to form a data record map, such records being primary data records having a direct connection with the target record and secondary data records being data records having a direct connection with a primary data record of the target record, wherein each primary and secondary data record is distinguished from the target record;

for at least one of the secondary data records, determining an importance value and a relevance value,

wherein the importance value of each secondary data record is determined based on:

interconnections between the secondary data record and other data records in the data record map; and

the predetermined data record strength value of each of the other data records having an interconnection with the secondary data record, and

wherein the relevance value of each secondary data record is determined based on:

interconnections between the secondary data records and other data records in the data record map relative to the target record; and

combining the importance value and the relevance value to provide the similarity score between the secondary data record and the target data record, wherein the similarity score allows a user to identify secondary data records that disclose similar concepts to the target data record.

2. A computer implemented method in accordance with claim 1 , wherein the importance value for each of the at least one secondary data records is determined by:

calculating a connection strength between the secondary data record and each primary data record connected to the secondary data record in the data map;

determining the predetermined data record strength value of each of the connected primary data records;

calculating a connection strength between the secondary data record with other secondary data records in the data map; and

determining the data record strength value of each of the secondary data records in the data set;

wherein the connection strengths and predetermined data record strength values are combined to form the importance value.

3. A computer implemented method in accordance with claim 1 , wherein the relevance value for each of the at least one secondary data records is determined by;

counting the number of times the secondary data record is connected to primary data records in the data map to form a first vote; and

counting the number of times the secondary data record is connected another secondary data record in the data map to form a second vote;

wherein the first and second votes are combined to form the relevance value.

4. A computer implemented method in accordance with claim 2 , wherein the relevance value for each of the at least one secondary data records is determined by;

calculating a connection distance between the secondary data records and the target data record in dependence on the connection strengths.

5. A computer implemented method in accordance with claim 1 , wherein the target data record is identified by a user.

6. A computer implemented method in accordance with claim 1 , wherein the importance value and the relevance value are attributed a weighting and combined to calculate the similarity score.

7. A computer implemented method in accordance with claim 1 , wherein the data records are patents, the primary data record being connected to the target data record in that they share a citation.

8. A computer implemented method in accordance with claim 1 , wherein the secondary data records are connected to the primary data records in that they share a citation.

9. A computer implemented method in accordance with claim 1 , wherein a visual representation of the data map shows connections between data records.

10. A computer implemented method according to claim 9 wherein the visual display includes phantom nodes, the phantom nodes being associated data records that are indirectly connected by links to the identified data record, and representing data records having a score below a predetermined threshold are represented, as are links with said phantom nodes, at a lower display intensity than data records having a score at or above the threshold.