### Table 1. Classi cation of trust functions

2004

"... In PAGE 34: ...Node in graph Edge in graph Web page ranks [10] Web page Hyperlink Personalized Web page ranks [7,6] Web page Hyperlink Document ranks (distributed) [12,14] Document on a peer Hyperlink between two documents Personalized document ranks (distributed) [1] Document on a peer Hyperlink between two documents Reputation values (distributed) [8] Peer Download experience of peers Personalized reputation values (distributed) - this paper Peer Download experience of peers Table1 . Hyperlink structures used by different ranking algorithms [7] introduces personalized PageRank Vectors (PPV) computed for each user.... In PAGE 66: ... Again, a subsumption may be supported by multiple nodes. Table1 and Table 2 show what these tables look like for the semantic network displayed in Fig. 3.... In PAGE 66: ... 3. Table1 . Class Instance Tables Table 2.... ..."

Cited by 5

### Table 1. PageRanks for the pages in Fig 3 Page Rank

"... In PAGE 5: ...o the cycle (B.C.D.E.F.G). This graph is treated as the web graph and the (unnormalized) PageRanks [6] of these nodes are shown below ( Table1 ) (PageRanks computed at http://www.... ..."

### Table 1: Graph theoretic correspondence between the Webgraph and the Trust Social Network.

"... In PAGE 3: ... Spammers, like social propagandists, form struc- tures that are able to gather a large number of such votes of confidence by design, thus breaking the assumption of independence in a hyperlink. Table1 has the correspondence between graph theoretic terms, the web graph according to a search engine, and the trust social network of a particular user. 4.... ..."

### Table 2: Ranking of usable paths for the paths digraph of Figure 7. rank 1 rank 2 rank 3 rank 4 rank 5 rank 6 rank 7 rank 8

"... In PAGE 11: ... The ranks of paths of G can be determined recursively, noting that the paths of rank j are those that have no edge pointing to them in the graph obtained from G by withdrawing all paths with rank less than j. Table2 shows the ranking of the usable paths of Figure 7. Let M be the rank of the highest rank path in G.... ..."

### Table 4: Our results for real web graph.

"... In PAGE 12: ... We can see that each update requires far fewer computations than would be needed for the full 61 million pages and 259 million edges. At the end of all the updates, we have the results described in Table4 . If we let be the original PageRank, ~ be the new correct PageRank, and ^ the approximation given by our algorithm, we have that k ? ~ k = 0:12056 while k^ ? ~ k = 5:9552 10?5, meaning that we apos;ve eliminated 99.... ..."

### Table 2. Predictors of Trust

"... In PAGE 6: ...nd yielded an R of .37 and a R Square of .136. As shown in Table2 , predictor variables that achieved significance were self-reported computer knowledge, self-reported Internet knowledge, and self-efficacy. Somewhat surprisingly, computer knowledge was negatively related to Trust.... ..."

### Table 7: Ranking performance for different trust information.

2006

"... In PAGE 18: ... We denotes these different combinations by CR(TR+PR), CR(Mass+PR) and CR(OTR+PR). Results in Table7 show that using the trust estimates generated by OTR achieves the best performance, corresponding to the top placement of OTR in the separation of spam and good pages from Section 4. 6.... ..."

### Table 2. Number of trust paths found for scale free topology for different TTL values. Target is 4 hops away.

"... In PAGE 7: ... Discovery of the first trust path. Table 1 and Table2 show the number of trust paths found, for different values of TTL, for the random and scale-free topology, respectively. The graphs in Figure 2 and Figure 3 show the number of messages propagated through the network for each algorithm, for different values of TTL.... In PAGE 7: ... Moreover, these messages can be redundant, because they forward a query to a node that earlier received that same query (see [14] for an in-depth discussion). However, Table 1 and Table2 demonstrate the benefit of this higher number of messages in the scale-free topology: the trust path is found within only a few hops for all algorithms. More precisely, the trust path is found in the minimum number of hops (TTL=3 for a target 4 hops away), except for K-walker with 10% forwarding.... ..."

### Table 1. Number of trust paths found for random graph topology for different TTL values. Target is 7 hops away. The last three lines indicate required TTL value to find at least one path but even with big TTL values Selective 10% did not find any path.

"... In PAGE 7: ... Discovery of the first trust path. Table1 and Table 2 show the number of trust paths found, for different values of TTL, for the random and scale-free topology, respectively. The graphs in Figure 2 and Figure 3 show the number of messages propagated through the network for each algorithm, for different values of TTL.... In PAGE 7: ... Moreover, these messages can be redundant, because they forward a query to a node that earlier received that same query (see [14] for an in-depth discussion). However, Table1 and Table 2 demonstrate the benefit of this higher number of messages in the scale-free topology: the trust path is found within only a few hops for all algorithms. More precisely, the trust path is found in the minimum number of hops (TTL=3 for a target 4 hops away), except for K-walker with 10% forwarding.... In PAGE 7: ... The minimum possible value of the TTL is 6 for a target at the average distance of 7 hops, but algorithms that do not forward to a high number of neighbours require higher TTL values. Table1 shows this number in the three last rows: to find the first trust path K-walker with 50% requires a TTL value of 11, and K-walker wit 10% needs TTL value 32. Moreover, selective forwarding with 10% never reaches the target, as indicated in Table 1.... ..."

### Table 1. Summary of large web page information

"... In PAGE 6: ... The pages were ranked using a simple three-point scale of Excellent, Good or Poor. Table1 summarizes the Web page information and presents the ranking for each page. Figure 6.... ..."