Activity
-
There are 2 MLE openings to work on large language model infra. An excellent opportunity to collaborate with top talents. Interested? Apply directly…
There are 2 MLE openings to work on large language model infra. An excellent opportunity to collaborate with top talents. Interested? Apply directly…
Liked by Haishan Liu
-
🚀 We are thrilled to announce the beta launch of our platform! 🚀 Our mission is to democratize investment opportunities, making pre-IPO investments…
🚀 We are thrilled to announce the beta launch of our platform! 🚀 Our mission is to democratize investment opportunities, making pre-IPO investments…
Liked by Haishan Liu
-
What an incredible experience to be part of the team that gets to launch the S2 premiere of #HouseOfTheDragon on Max. Huge shoutout to everyone…
What an incredible experience to be part of the team that gets to launch the S2 premiere of #HouseOfTheDragon on Max. Huge shoutout to everyone…
Liked by Haishan Liu
Experience
Education
Publications
-
Generating supplemental content information using virtual profiles
ACM RecSys 2013
We describe a hybrid recommendation system at LinkedIn that seeks to optimally extract relevant information pertaining to items to be recommended. By extending the notion of an item profile, we propose the concept of a "virtual profile" that augments the content of the item with rich set of features inherited from members who have already shown explicit interest in it. Unlike item-based collaborative filtering, we focus on discovering the characteristic descriptors that underlie the item-user…
We describe a hybrid recommendation system at LinkedIn that seeks to optimally extract relevant information pertaining to items to be recommended. By extending the notion of an item profile, we propose the concept of a "virtual profile" that augments the content of the item with rich set of features inherited from members who have already shown explicit interest in it. Unlike item-based collaborative filtering, we focus on discovering the characteristic descriptors that underlie the item-user association. Such information is used as supplemental features in a content-based filtering system. The main objective of virtual profiles is to provide a means to tap into rich-content information from one type of entity and propagate features extracted from which to other affiliated entities that may suffer from relative data scarcity. We empirically evaluate the proposed method on a real-world community recommendation problem at LinkedIn. The result shows that the virtual profiles outperform a collaborative filtering based approach (user who likes this also likes that). In particular, the improvement is more significant for new users with only limited connections, demonstrating the capability of the method to address the cold-start problem in pure collaborative filtering systems.
Other authorsSee publication -
Breaking the Deadlock: Simultaneously Discovering Attribute Matching and Cluster Matching with Multi-Objective Metaheuristics
Journal on Data Semantics: Volume 1, Issue 2 (2012), Page 133-145, DOI: 10.1007/s13740-012-0010-0
In this paper, we present a data mining approach to address challenges in the matching of heterogeneous datasets. In particular, we propose solutions to two problems that arise in integrating information from different results of scientific research. The first problem, attribute matching, involves discovery of correspondences among distinct numeric features (attributes) that are used to characterize datasets that have been collected and analyzed in different research labs. The second problem…
In this paper, we present a data mining approach to address challenges in the matching of heterogeneous datasets. In particular, we propose solutions to two problems that arise in integrating information from different results of scientific research. The first problem, attribute matching, involves discovery of correspondences among distinct numeric features (attributes) that are used to characterize datasets that have been collected and analyzed in different research labs. The second problem, cluster matching, involves discovery of matchings between patterns (clusters) across datasets. We treat both of these problems together as a multi-objective optimization problem. A multi-objective simulated annealing algorithm is described to find the optimal solution and compared with the genetic algorithm. The utility of this approach is demonstrated in a series of experiments using synthetic and realistic datasets that are designed to simulate heterogeneous data from different sources.
Other authorsSee publication -
Sharing and Integration of cognitive neuroscience data: metric and pattern matching across heterogeneous ERP datasets
Neurocomputing (2012), doi:10.1016/j.neucom.2012.01.028
-
A Hypergraph-based Method for Discovering Semantically Associated Itemsets
In Proceedings of the 11th IEEE International Conference on Data Mining (ICDM)
In this paper, we address an interesting data mining problem of finding semantically associated item sets, i.e., items connected via indirect links. We propose a novel method for discovering semantically associated item sets based on a hyper graph representation of the database. We describe two similarity measures to compute the strength of associations between items. Specifically, we introduce the average commute time similarity, $\mathbf{s_{CT}}$, based on the random walk model on hyper…
In this paper, we address an interesting data mining problem of finding semantically associated item sets, i.e., items connected via indirect links. We propose a novel method for discovering semantically associated item sets based on a hyper graph representation of the database. We describe two similarity measures to compute the strength of associations between items. Specifically, we introduce the average commute time similarity, $\mathbf{s_{CT}}$, based on the random walk model on hyper graph, and the inner-product similarity, $\mathbf{s_{L+}}$, based on the Moore-Penrose pseudoinverse of the hyper graph Laplacian matrix. Given semantically associated 2-itemsets generated by these measures, we design a hyper graph expansion method with two search strategies, namely, the clique and connected component search, to generate $k$-item sets ($k>2$). We show the proposed method is indeed capable of capturing semantically associated item sets through experiments performed on three datasets ranging from low to high dimensionality. The semantically associated item sets discovered in our experiment is promising to provide valuable insights on interrelationship between medical concepts and other domain specific concepts.
Other authorsSee publication -
Breaking the Deadlock: Simultaneously Discovering Attribute Matching and Cluster Matching with Multi-Objective Simulated Annealing
In Proceedings of the International Conference on Ontologies, Databases and Application of SEmantics (ODBASE 2011). LNCS 7045, pp. 698-715,
In this paper, we present a data mining approach to chal- lenges in the matching and integration of heterogeneous datasets. In particular, we propose solutions to two problems that arise in combining information from different results of scientific research. The first problem, attribute matching, involves discovery of correspondences among distinct numeric-typed summary features (“attributes”) that are used to charac- terize datasets that have been collected and analyzed in different research…
In this paper, we present a data mining approach to chal- lenges in the matching and integration of heterogeneous datasets. In particular, we propose solutions to two problems that arise in combining information from different results of scientific research. The first problem, attribute matching, involves discovery of correspondences among distinct numeric-typed summary features (“attributes”) that are used to charac- terize datasets that have been collected and analyzed in different research labs. The second problem, cluster matching, involves discovery of matchings between patterns across datasets. We treat both of these problems together as a multi-objective optimization problem. A multi-objective simulated annealing algorithm is described to find the optimal solution. The utility of this approach is demonstrated in a series of experiments using synthetic and realistic datasets that are designed to simulate heterogeneous data from different sources.
Other authorsSee publication -
Ontology-based Mining of Brainwaves: A Sequence Similarity Technique for Mapping Alternative Descriptions of Patterns in Event Related Potentials (ERP) Data
In Proceedings of the 14th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2010). LNCS 6119, pp. 43-54
-
Towards Semantic Data Mining
In Doctoral Consortium of the 9th International Semantic Web Conference (ISWC)
Incorporating domain knowledge is one of the most challenging problems in data mining. The Semantic Web technologies are promising to offer solutions to formally capture domain knowledge and realize the efficient usage of the knowledge. We call the the data mining technology powered by the Semantic Web, capable of systematically incorporating domain knowledge, the semantic data mining. In this paper, we first survey a body of representative academic work that explores ontology- based…
Incorporating domain knowledge is one of the most challenging problems in data mining. The Semantic Web technologies are promising to offer solutions to formally capture domain knowledge and realize the efficient usage of the knowledge. We call the the data mining technology powered by the Semantic Web, capable of systematically incorporating domain knowledge, the semantic data mining. In this paper, we first survey a body of representative academic work that explores ontology- based optimization for various data mining applications. Then we identify the semantic annotation as a crucial step towards semantic data mining since it brings meaning to data. Finally we propose a learning- based semantic search algorithm for annotating (semi-) structured data.
-
An Exploration of Understanding Heterogeneity through Data Mining
In Proceedings of KDD'08 Workshop on Mining Multiple Information Sources (MMIS 2008). pp. 18-25
Patents
More activity by Haishan
-
Come join us to build the future of CTV advertising! https://lnkd.in/dVPDN3sp
Come join us to build the future of CTV advertising! https://lnkd.in/dVPDN3sp
Liked by Haishan Liu
-
I am deeply honored and humbled to receive the ICDE 10-Year Influential Paper Award. I wish I could have been there in person! Graph partitioning…
I am deeply honored and humbled to receive the ICDE 10-Year Influential Paper Award. I wish I could have been there in person! Graph partitioning…
Liked by Haishan Liu
-
My first RSA conference. It was wonderful to see so many in the industry working on the tough cybersecurity challenges to keep our world safe. It was…
My first RSA conference. It was wonderful to see so many in the industry working on the tough cybersecurity challenges to keep our world safe. It was…
Liked by Haishan Liu
-
Thank you Swiss Financial Innovation Desk, I am really appreciative of being able to contribute to this effort and encourage innovation in this…
Thank you Swiss Financial Innovation Desk, I am really appreciative of being able to contribute to this effort and encourage innovation in this…
Liked by Haishan Liu
-
I am very excited to announce that I have joined Cohesity as their Chief Technology Officer. I have had strong relationship with Cohesity for many…
I am very excited to announce that I have joined Cohesity as their Chief Technology Officer. I have had strong relationship with Cohesity for many…
Liked by Haishan Liu
-
Congratulations to Ken Doctor and the crew of Lookout Santa Cruz for winning the Pulitzer for breaking news. Wow. Just … wow. https://lnkd.in/gQQN4dEG
Congratulations to Ken Doctor and the crew of Lookout Santa Cruz for winning the Pulitzer for breaking news. Wow. Just … wow. https://lnkd.in/gQQN4dEG
Liked by Haishan Liu
-
Today is my last day at SmartNews! It's been an exhilarating, challenging and rewarding 3 years building our ad strategy in the US from the ground…
Today is my last day at SmartNews! It's been an exhilarating, challenging and rewarding 3 years building our ad strategy in the US from the ground…
Liked by Haishan Liu
-
It's hard to say goodbye after 12 years of working at LinkedIn, where I have felt as much at home as home. Through teamwork, we've risen to…
It's hard to say goodbye after 12 years of working at LinkedIn, where I have felt as much at home as home. Through teamwork, we've risen to…
Liked by Haishan Liu
-
To all my LinkedIn friends impacted today - please reach out if you are looking for a new opportunity. I have a number of openings in my org in ML…
To all my LinkedIn friends impacted today - please reach out if you are looking for a new opportunity. I have a number of openings in my org in ML…
Liked by Haishan Liu
-
Thrilled to be stepping into the role of Chief People Officer at “People First” company. Grateful for this incredible opportunity!
Thrilled to be stepping into the role of Chief People Officer at “People First” company. Grateful for this incredible opportunity!
Liked by Haishan Liu
Other similar profiles
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore MoreOthers named Haishan Liu in United States
-
Haishan Liu
Actively Seeking for MLE/DS/RS internship | ECE PhD Candidate | Deep Learning | Reinforcement Learning | Order Dispatching
-
Haishan Liu
--
-
Haishan Liu
Marketing Intern
-
刘海珊
The Johns Hopkins University学生
4 others named Haishan Liu in United States are on LinkedIn
See others named Haishan Liu