TY - GEN
T1 - X-Rank
T2 - 27th ACM International Conference on Information and Knowledge Management, CIKM 2018
AU - Kang, Jian
AU - Xia, Yinglong
AU - Freitas, Scott
AU - Cao, Nan
AU - Yu, Haichao
AU - Tong, Hanghang
N1 - Funding Information:
Results. Findings from the user study are reported in Table 1. From this table we offer the following observa- tions: (1) in 5 out of 6 queries, X-Rank performs the best among all compared meth- ods; (2) comparing X-Rank and CrossQuery, the signif- icant improvement indicates Figure 4: Explainability rat-theeffectivenessofadding ings of all compared methods. knowledge layers. Further- Higherisbetter. more, when measuring the usefulness of explanations from X-Rank, the users gave the explanations an average rating of 4.22 (out of 5). This is significantly higher than the ratings for RWR, HITS and CrossQuery—which are 3.60, 3.55, 3.31, respectively. Results are shown in Figure 4. This demonstrates the potential of the proposed X-Rank algorithm to provide useful and intuitive explanations. In addition, the X-Rank algorithm is capable of scaling to large networks due to its linear complexity. To see this, we note that LocalProximity, CrossQuery, Aurora-E and Aurora-N all have linear complexities, which renders a linear time complexity of the overall X-Rank algorithm. 5 CONCLUSION The goal of this work is to develop a web-based prototype (X-Rank) for researchers and practitioners to visually explore and interact with the proposed explainable ranking algorithm. We believe the platform and algorithm will be of particular interest to both researchers and practitioners in the fields of information retrieval and data mining. In addition, an operational prototype of the X-Rank platform is currently online (http://www.x-rank.net), along with a demonstration video (https://youtu.be/EAKPaCWJQxQ). Source code will be made publicly available by the conference date. ACKNOWLEDGMENTS This work is supported by National Science Foundation under Grant No. IIS-1651203, IIS-1715385, CNS-1629888 and IIS-1743040, DTRA under the grant number HDTRA1-16-0017, Army Research Office under the contract number W911NF-16-1-0168, Department of Homeland Security under Grant Award Number 2017-ST-061-QA0001, National Natural Science Foundation of China under the grant number 61602306, and gifts from Huawei and Baidu. REFERENCES
Funding Information:
This work is supported by National Science Foundation under Grant No. IIS-1651203, IIS-1715385, CNS-1629888 and IIS-1743040, DTRA under the grant number HDTRA1-16-0017, Army Research Office under the contract number W911NF-16-1-0168, Department of Homeland Security under Grant Award Number 2017-ST-061-QA0001, National Natural Science Foundation of China under the grant number 61602306, and gifts from Huawei and Baidu.
Publisher Copyright:
© 2018 Association for Computing Machinery.
PY - 2018/10/17
Y1 - 2018/10/17
N2 - In this paper we present a web-based prototype for an explainable ranking algorithm in multi-layered networks, incorporating both network topology and knowledge information. While traditional ranking algorithms such as PageRank and HITS are important tools for exploring the underlying structure of networks, they have two fundamental limitations in their efforts to generate high accuracy rankings. First, they are primarily focused on network topology, leaving out additional sources of information (e.g. attributes, knowledge). Secondly, most algorithms do not provide explanations to the end-users on why the algorithm gives the specific ranking results, hindering the usability of the ranking information. We developed X-Rank, an explainable ranking tool, to address these drawbacks. Empirical results indicate that our explainable ranking method not only improves ranking accuracy, but facilitates user understanding of the ranking by exploring the top influential elements in multi-layered networks. The web-based prototype (X-Rank: http://www.x-rank.net) is currently online-we believe it will assist both researchers and practitioners looking to explore and exploit multi-layered network data.
AB - In this paper we present a web-based prototype for an explainable ranking algorithm in multi-layered networks, incorporating both network topology and knowledge information. While traditional ranking algorithms such as PageRank and HITS are important tools for exploring the underlying structure of networks, they have two fundamental limitations in their efforts to generate high accuracy rankings. First, they are primarily focused on network topology, leaving out additional sources of information (e.g. attributes, knowledge). Secondly, most algorithms do not provide explanations to the end-users on why the algorithm gives the specific ranking results, hindering the usability of the ranking information. We developed X-Rank, an explainable ranking tool, to address these drawbacks. Empirical results indicate that our explainable ranking method not only improves ranking accuracy, but facilitates user understanding of the ranking by exploring the top influential elements in multi-layered networks. The web-based prototype (X-Rank: http://www.x-rank.net) is currently online-we believe it will assist both researchers and practitioners looking to explore and exploit multi-layered network data.
KW - Explainability
KW - Knowledge
KW - Multi-layered network
KW - Ranking
UR - http://www.scopus.com/inward/record.url?scp=85058058915&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85058058915&partnerID=8YFLogxK
U2 - 10.1145/3269206.3269224
DO - 10.1145/3269206.3269224
M3 - Conference contribution
AN - SCOPUS:85058058915
T3 - International Conference on Information and Knowledge Management, Proceedings
SP - 1959
EP - 1962
BT - CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management
A2 - Paton, Norman
A2 - Candan, Selcuk
A2 - Wang, Haixun
A2 - Allan, James
A2 - Agrawal, Rakesh
A2 - Labrinidis, Alexandros
A2 - Cuzzocrea, Alfredo
A2 - Zaki, Mohammed
A2 - Srivastava, Divesh
A2 - Broder, Andrei
A2 - Schuster, Assaf
PB - Association for Computing Machinery
Y2 - 22 October 2018 through 26 October 2018
ER -