Mining a Maximum Weighted Set of Disjoint Submatrices

Branders, Vincent; Derval, Guillaume; Schaus, Pierre; Dupont, Pierre

doi:10.1007/978-3-030-33778-0_2

Cited by 3 publications

(2 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Max-sum submatrix [9,10] targets highly expressed subsets of genes and of samples by identifying a submatrix with the maximum sum. [17] finds 𝑘 submatrices by constraint programming and [8] adds the submatrix disjointness constraint. While similar in maximizing the sum of entries, there are key differences between these works and ours.…”

Section: Related Workmentioning

confidence: 99%

Summarizing User-item Matrix By Group Utility Maximization

Wang

Long

et al. 2023

ACM Trans. Knowl. Discov. Data

View full text Add to dashboard Cite

A user-item utility matrix represents the utility (or preference) associated with each (user, item) pair, such as citation counts, rating/vote on items or locations, and clicks on items. A high utility value indicates a strong association of the pair. In this work, we consider the problem of summarizing strong association for a large user-item matrix using a small summary size. Traditional techniques fail to distinguish user groups associated with different items (such as top- l item selection) or fail to focus on high utility (such as similarity based subspace clustering and biclustering). We formulate a new problem, called Group Utility Maximization, to summarize the entire user population through k user groups and l items for each group; the goal is to maximize the total utility of selected items over all groups collectively. We show this problem is NP-hard even for l = 1. We present two algorithms. One greedily finds the next group, called Greedy algorithm, and the other iteratively refines existing k groups, called k -max algorithm. Greedy algorithm provides the \((1-\frac{1}{e}) \) approximation guarantee for a nonnegative utility matrix, whereas k -max algorithm is more efficient for large datasets. We evaluate these algorithms on real-life datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

Summarizing User-item Matrix By Group Utility Maximization

Wang

Long

et al. 2023

ACM Trans. Knowl. Discov. Data

View full text Add to dashboard Cite

show abstract

“…Max-sum submatrix [142][143] targets highly expressed subsets of genes and of samples by identifying a submatrix with the maximum sum. [135] finds k submatrices by constraint programming and [144] adds the submatrix disjointness constraint. While similar in maximizing the sum of entries, there are key di↵erences between these works and ours.…”

Section: Related Workmentioning

confidence: 99%

Counterfactual explanations for machine learning models on heterogeneous data

Wang

View full text Add to dashboard Cite

Miao, for her professional and insightful guidance throughout my four-year research. She taught me to explore a new research field, write scientific papers, and most importantly, become an independent researcher. I would also like to express my sincerest appreciation to Professor Ke Wang and Professor Cheng Long. Their patience, motivation, enthusiasm, and knowledge have supported me through all stages of my Ph.D. Without their help, I would not have been able to complete my Ph.D. I would also like to thank my Ph.D. committee members for their thoughtful comments and constructive suggestions.Second, I would like to extend my gratitude to my fellow labmates at NTU: Hangwei Qian, Qinxu Ding, and Yong Liu for discussing and revising my papers during many sleepless nights before the deadlines. I am also grateful to have had lovely badminton and basketball buddies:

show abstract

Summarizing User-Item Matrix By Group Utility Maximization

Wang

Long

et al. 2021

2021 IEEE International Conference on Data Mining (ICDM)

View full text Add to dashboard Cite

A user-item utility matrix represents the utility (or preference) associated with each (user, item) pair, such as citation counts, rating/vote on items or locations, and clicks on items. A high utility value indicates a strong association of the pair. In this work, we consider the problem of summarizing strong association for a large user-item matrix using a small summary size. Traditional techniques fail to distinguish user groups associated with different items (such as top-𝑙 item selection) or fail to focus on high utility (such as similarity based subspace clustering and biclustering). We formulate a new problem, called Group Utility Maximization, to summarize the entire user population through 𝑘 user groups and 𝑙 items for each group; the goal is to maximize the total utility of selected items over all groups collectively. We show this problem is NP-hard even for 𝑙 = 1. We present two algorithms. One greedily finds the next group, called Greedy algorithm, and the other iteratively refines existing 𝑘 groups, called 𝑘-max algorithm. Greedy algorithm provides the (1 − 1 𝑒 ) approximation guarantee for a nonnegative utility matrix, whereas 𝑘-max algorithm is more efficient for large datasets. We evaluate these algorithms on real-life datasets.

show abstract

Mining a Maximum Weighted Set of Disjoint Submatrices

Cited by 3 publications

References 10 publications

Summarizing User-item Matrix By Group Utility Maximization

Summarizing User-item Matrix By Group Utility Maximization

Counterfactual explanations for machine learning models on heterogeneous data

Summarizing User-Item Matrix By Group Utility Maximization

Contact Info

Product

Resources

About