Semi-supervised protein classification using cluster kernels

Weston, Jason; Leslie, Christina; Ie, Eugene; Zhou, Dengyong; Elisseeff, André; Noble, William Stafford

doi:10.1093/bioinformatics/bti497

Cited by 184 publications

(198 citation statements)

References 18 publications

Supporting

Mentioning

196

Contrasting

Unclassified

Order By: Relevance

“…Previous approaches include using cluster structure for predicting class labels on graphs [26], and using cluster kernels for semi-supervised classification [27]. Our approach is scalable to large graphs due to the scalability of the underlying role discovery approach, and outperforms other state-of-the-art algorithms in classification tasks over graphs.…”

Section: Related Workmentioning

confidence: 99%

RolX: Role Extraction and Mining in Large Networks

Henderson¹,

Gallagher²,

Eliassi-Rad³

et al. 2011

204

287

View full text Add to dashboard Cite

Abstract-Given a graph, how can we automatically discover roles for nodes? Roles could be, eg., 'bridges', or 'peripherynodes', etc. Roles are compact summaries of a node's behavior that generalize across networks. They enable numerous novel and useful network mining tasks, such as sense-making, searching for similar nodes, and node classification. We propose RolX (Role eXtraction), a scalable (linear in the number of edges), unsupervised learning approach for automatically extracting roles from general network data. We demonstrate the effectiveness of RolX on several network mining tasks, from exploratory data analysis to network transfer learning. Moreover, we compare network role discovery with network community discovery. We highlight fundamental differences between the two (e.g., roles generalize across disconnected networks, communities do not).

show abstract

Section: Related Workmentioning

confidence: 99%

RolX: Role Extraction and Mining in Large Networks

Henderson¹,

Gallagher²,

Eliassi-Rad³

et al. 2011

204

287

View full text Add to dashboard Cite

show abstract

“…Therefore, the behavior of a user's close friends are also informative in identifying different accounts of the same user. In [34], the average similarity of the neighborhood data of two data items is more robust compared with the original similarity since it calculates the similarity of two convex hulls instead of two data points. Inspired by [34], we model the behavior of a user's social connections.…”

Section: Core Social Network Featuresmentioning

confidence: 99%

“…In [34], the average similarity of the neighborhood data of two data items is more robust compared with the original similarity since it calculates the similarity of two convex hulls instead of two data points. Inspired by [34], we model the behavior of a user's social connections. Given two users i and i from different platforms, the behavior data of their top-k most frequently interacting friends are collected.…”

Section: Core Social Network Featuresmentioning

confidence: 99%

Structured Learning from Heterogeneous Behavior for Social Identity Linkage

Liu

Wang

Zhu

2015

IEEE Trans. Knowl. Data Eng.

View full text Add to dashboard Cite

Abstract-Social identity linkage across different social media platforms is of critical importance to business intelligence by gaining from social data a deeper understanding and more accurate profiling of users. In this paper, we propose a solution framework, HYDRA, which consists of three key steps: (I) we model heterogeneous behavior by long-term topical distribution analysis and multi-resolution temporal behavior matching against high noise and information missing, and the behavior similarity are described by multi-dimensional similarity vector for each user pair; (II) we build structure consistency models to maximize the structure and behavior consistency on users' core social structure across different platforms, thus the task of identity linkage can be performed on groups of users, which is beyond the individual level linkage in previous study; and (III) we propose a normalized-margin-based linkage function formulation, and learn the linkage function by multi-objective optimization where both supervised pair-wise linkage function learning and structure consistency maximization are conducted towards a unified Pareto optimal solution. The model is able to deal with drastic information missing, and avoid the curse-of-dimensionality in handling high dimensional sparse representation. Extensive experiments on 10 million users across seven popular social networks platforms demonstrate that HYDRA correctly identifies real user linkage across different platforms from massive noisy user behavior data records, and outperforms existing state-of-the-art approaches by at least 20% under different settings, and 4 times better in most settings.

show abstract

“…Among the discriminative approaches, string kernel-based machine learning methods provide some of the most accurate results [27,19,16,28].…”

Section: String Kernelsmentioning

confidence: 99%

“…In terms of semi-supervised extensions of string kernels, another very simple method, called the "sequence neighborhood" kernel or "cluster" kernel has been employed [28] previously. This method replaces every example with a new representation obtained by averaging representations of the example's neighbors found in the unlabeled data using some standard sequence similarity measure.…”

Section: Semi-supervised String Kernelmentioning

confidence: 99%

Semi-supervised Abstraction-Augmented String Kernel for Multi-level Bio-Relation Extraction

Kuksa

Bai

et al. 2010

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Bio-relation extraction (bRE), an important goal in bio-text mining, involves subtasks identifying relationships between bio-entities in text at multiple levels, e.g., at the article, sentence or relation level. A key limitation of current bRE systems is that they are restricted by the availability of annotated corpora. In this work we introduce a semisupervised approach that can tackle multi-level bRE via string comparisons with mismatches in the string kernel framework. Our string kernel implements an abstraction step, which groups similar words to generate more abstract entities, which can be learnt with unlabeled data. Specifically, two unsupervised models are proposed to capture contextual (local or global) semantic similarities between words from a large unannotated corpus. This Abstraction-augmented String Kernel (ASK) allows for better generalization of patterns learned from annotated data and provides a unified framework for solving bRE with multiple degrees of detail. ASK shows effective improvements over classic string kernels on four datasets and achieves state-of-the-art bRE performance without the need for complex linguistic features.

show abstract

Semi-supervised protein classification using cluster kernels

Abstract: www.kyb.tuebingen.mpg.de/bs/people/weston/semiprot.

Cited by 184 publications

References 18 publications

RolX: Role Extraction and Mining in Large Networks

RolX: Role Extraction and Mining in Large Networks

Structured Learning from Heterogeneous Behavior for Social Identity Linkage

Semi-supervised Abstraction-Augmented String Kernel for Multi-level Bio-Relation Extraction

Contact Info

Product

Resources

About