PathSim: meta path-based top-K similarity search in heterogeneous information networks
Citations
2,494 citations
[...]
1,934 citations
1,794 citations
Cites background or methods from "PathSim: meta path-based top-K simi..."
...Table 4 shows the node clustering results as measured by NMI in the AMiner CS data....
[...]
...We summarize the dierences of these methods in Table 1, which lists their input to learning algorithms, as well as the top-ve similarity search results in the DBIS network for the same two queries used in [26] (see Section 4 for details)....
[...]
...demonstrated that heterogeneous random walks are biased to highly visible types of nodes—those with a dominant number of paths—and concentrated nodes—those with a governing percentage of paths pointing to a small set of nodes [26]....
[...]
...Method PathSim [26] DeepWalk / node2vec [8, 22] LINE (1st+2nd) [30] PTE [29] metapath2vec metapath2vec++...
[...]
...Here we leverage the k-means algorithm to cluster the data and evaluate the clustering results in terms of normalized mutual information (NMI) [26]....
[...]
1,467 citations
1,266 citations
References
13,789 citations
11,827 citations
"PathSim: meta path-based top-K simi..." refers methods in this paper
...We apply Normalized Cut [15] to the 3 similarity matrices, and use NMI (Normalized Mutual Information, with the value between 0 and 1, the higher the better) [16] to calculate the clustering accuracy for both venues and authors, and their weighted average accuracy over the two types....
[...]
10,500 citations
4,337 citations
"PathSim: meta path-based top-K simi..." refers result in this paper
...Then we use the measure nDCG (Normalized Discounted Cumulative Gain, with the value between 0 and 1, the higher the better) [9] to evaluate the quality of a ranking algorithm by comparing its output ranking results with the labeled ones (Table 5)....
[...]
2,036 citations