Merge pull request #33 from WING-NUS/hugoblox-import-publications

Hugo Blox Builder - Import latest publications

content/publication/10-1002-asi-10364/cite.bib

@@ -0,0 +1,17 @@
+@article{10.1002/asi.10364,
+ address = {USA},
+ author = {Kan, Min-Yen},
+ doi = {10.1002/asi.10364},
+ issn = {1532-2882},
+ issue_date = {January 15, 2004},
+ journal = {J. Am. Soc. Inf. Sci. Technol.},
+ month = {January},
+ number = {2},
+ numpages = {2},
+ pages = {178–179},
+ publisher = {John Wiley & Sons, Inc.},
+ title = {Review of "Introduction to digital libraries by G. G. Chowdhury and Sudatta Chowdhury" London: Facet 2003},
+ url = {https://doi.org/10.1002/asi.10364},
+ volume = {55},
+ year = {2004}
+}

content/publication/10-1002-asi-10364/index.md

@@ -0,0 +1,15 @@
+---
+title: 'Review of \"Introduction to digital libraries by G. G. Chowdhury and Sudatta
+  Chowdhury\" London: Facet 2003'
+authors:
+- Min-Yen Kan
+date: '2004-01-01'
+publishDate: '2024-10-20T00:58:41.643721Z'
+publication_types:
+- article-journal
+publication: '*J. Am. Soc. Inf. Sci. Technol.*'
+doi: 10.1002/asi.10364
+links:
+- name: URL
+  url: https://doi.org/10.1002/asi.10364
+---

content/publication/10-1007-11788034-15/cite.bib

@@ -0,0 +1,16 @@
+@inproceedings{10.1007/11788034_15,
+ abstract = {Recent research in video retrieval has focused on automated, high-level feature indexing on shots or frames. One important application of such indexing is to support precise video retrieval. We report on extensions of this semantic indexing on news video retrieval. First, we utilize extensive query analysis to relate various high-level features and query terms by matching the textual description and context in a time-dependent manner. Second, we introduce a framework to effectively fuse the relation weights with the detectors' confidence scores. This results in individual high level features that are weighted on a per-query basis. Tests on the TRECVID 2005 dataset show that the above two enhancements yield significant improvement in performance over a corresponding state-of-the-art video retrieval baseline.},
+ address = {Berlin, Heidelberg},
+ author = {Neo, Shi-Yong and Zhao, Jin and Kan, Min-Yen and Chua, Tat-Seng},
+ booktitle = {Proceedings of the 5th International Conference on Image and Video Retrieval},
+ doi = {10.1007/11788034_15},
+ isbn = {3540360182},
+ location = {Tempe, AZ},
+ numpages = {10},
+ pages = {143–152},
+ publisher = {Springer-Verlag},
+ series = {CIVR'06},
+ title = {Video retrieval using high level features: exploiting query matching and confidence-based weighting},
+ url = {https://doi.org/10.1007/11788034_15},
+ year = {2006}
+}

content/publication/10-1007-11788034-15/index.md

@@ -0,0 +1,30 @@
+---
+title: 'Video retrieval using high level features: exploiting query matching and confidence-based
+  weighting'
+authors:
+- Shi-Yong Neo
+- Jin Zhao
+- Min-Yen Kan
+- Tat-Seng Chua
+date: '2006-01-01'
+publishDate: '2024-10-20T00:58:41.603586Z'
+publication_types:
+- paper-conference
+publication: '*Proceedings of the 5th International Conference on Image and Video
+  Retrieval*'
+doi: 10.1007/11788034_15
+abstract: Recent research in video retrieval has focused on automated, high-level
+  feature indexing on shots or frames. One important application of such indexing
+  is to support precise video retrieval. We report on extensions of this semantic
+  indexing on news video retrieval. First, we utilize extensive query analysis to
+  relate various high-level features and query terms by matching the textual description
+  and context in a time-dependent manner. Second, we introduce a framework to effectively
+  fuse the relation weights with the detectors' confidence scores. This results in
+  individual high level features that are weighted on a per-query basis. Tests on
+  the TRECVID 2005 dataset show that the above two enhancements yield significant
+  improvement in performance over a corresponding state-of-the-art video retrieval
+  baseline.
+links:
+- name: URL
+  url: https://doi.org/10.1007/11788034_15
+---

content/publication/10-1007-11788034-48/cite.bib

@@ -0,0 +1,16 @@
+@inproceedings{10.1007/11788034_48,
+ abstract = {We introduce NPIC, an image classification system that focuses on synthetic (e.g., non-photographic) images. We use class-specific keywords in an image search engine to create a noisily labeled training corpus of images for each class. NPIC then extracts both content-based image retrieval (CBIR) features and metadata-based textual features for each image for machine learning. We evaluate this approach on three different granularities: 1) natural vs. synthetic, 2) map vs. figure vs. icon vs. cartoon vs. artwork 3) and further subclasses of the map and figure classes. The NPIC framework achieves solid performance (99%, 97% and 85% in cross validation, respectively). We find that visual features provide a significant boost in performance, and that textual and visual features vary in usefulness at the different levels of granularities of classification.},
+ address = {Berlin, Heidelberg},
+ author = {Wang, Fei and Kan, Min-Yen},
+ booktitle = {Proceedings of the 5th International Conference on Image and Video Retrieval},
+ doi = {10.1007/11788034_48},
+ isbn = {3540360182},
+ location = {Tempe, AZ},
+ numpages = {10},
+ pages = {473–482},
+ publisher = {Springer-Verlag},
+ series = {CIVR'06},
+ title = {NPIC: hierarchical synthetic image classification using image search and generic features},
+ url = {https://doi.org/10.1007/11788034_48},
+ year = {2006}
+}

content/publication/10-1007-11788034-48/index.md

@@ -0,0 +1,28 @@
+---
+title: 'NPIC: hierarchical synthetic image classification using image search and generic
+  features'
+authors:
+- Fei Wang
+- Min-Yen Kan
+date: '2006-01-01'
+publishDate: '2024-10-20T00:58:41.596622Z'
+publication_types:
+- paper-conference
+publication: '*Proceedings of the 5th International Conference on Image and Video
+  Retrieval*'
+doi: 10.1007/11788034_48
+abstract: 'We introduce NPIC, an image classification system that focuses on synthetic
+  (e.g., non-photographic) images. We use class-specific keywords in an image search
+  engine to create a noisily labeled training corpus of images for each class. NPIC
+  then extracts both content-based image retrieval (CBIR) features and metadata-based
+  textual features for each image for machine learning. We evaluate this approach
+  on three different granularities: 1) natural vs. synthetic, 2) map vs. figure vs.
+  icon vs. cartoon vs. artwork 3) and further subclasses of the map and figure classes.
+  The NPIC framework achieves solid performance (99%, 97% and 85% in cross validation,
+  respectively). We find that visual features provide a significant boost in performance,
+  and that textual and visual features vary in usefulness at the different levels
+  of granularities of classification.'
+links:
+- name: URL
+  url: https://doi.org/10.1007/11788034_48
+---

content/publication/10-1016-j-artmed-2004-07-018/cite.bib

@@ -0,0 +1,19 @@
+@article{10.1016/j.artmed.2004.07.018,
+ abstract = {Objective:: We present the summarization system in the PErsonalized Retrieval and Summarization of Images, Video and Language (PERSIVAL) medical digital library. Although we discuss the context of our summarization research within the PERSIVAL platform, the primary focus of this article is on strategies to define and generate customized summaries. Methods and material:: Our summarizer employs a unified user model to create a tailored summary of relevant documents for either a physician or lay person. The approach takes advantage of regularities in medical literature text structure and content to fulfill identified user needs. Results:: The resulting summaries combine both machine-generated text and extracted text that comes from multiple input documents. Customization includes both group-based modeling for two classes of users, physician and lay person, and individually driven models based on a patient record. Conclusions:: Our research shows that customization is feasible in a medical digital library.},
+ address = {GBR},
+ author = {Elhadad, N. and Kan, M. -Y. and Klavans, J. L. and McKeown, K. R.},
+ doi = {10.1016/j.artmed.2004.07.018},
+ issn = {0933-3657},
+ issue_date = {February, 2005},
+ journal = {Artif. Intell. Med.},
+ keywords = {Clinical information system, Medical digital library, Multi-document information extraction, Multi-document summarization, User modeling},
+ month = {February},
+ number = {2},
+ numpages = {20},
+ pages = {179–198},
+ publisher = {Elsevier Science Publishers Ltd.},
+ title = {Customization in a unified framework for summarizing medical literature},
+ url = {https://doi.org/10.1016/j.artmed.2004.07.018},
+ volume = {33},
+ year = {2005}
+}

content/publication/10-1016-j-artmed-2004-07-018/index.md

@@ -0,0 +1,36 @@
+---
+title: Customization in a unified framework for summarizing medical literature
+authors:
+- N. Elhadad
+- M. -Y. Kan
+- J. L. Klavans
+- K. R. McKeown
+date: '2005-02-01'
+publishDate: '2024-10-20T00:58:41.629291Z'
+publication_types:
+- article-journal
+publication: '*Artif. Intell. Med.*'
+doi: 10.1016/j.artmed.2004.07.018
+abstract: 'Objective:: We present the summarization system in the PErsonalized Retrieval
+  and Summarization of Images, Video and Language (PERSIVAL) medical digital library.
+  Although we discuss the context of our summarization research within the PERSIVAL
+  platform, the primary focus of this article is on strategies to define and generate
+  customized summaries. Methods and material:: Our summarizer employs a unified user
+  model to create a tailored summary of relevant documents for either a physician
+  or lay person. The approach takes advantage of regularities in medical literature
+  text structure and content to fulfill identified user needs. Results:: The resulting
+  summaries combine both machine-generated text and extracted text that comes from
+  multiple input documents. Customization includes both group-based modeling for two
+  classes of users, physician and lay person, and individually driven models based
+  on a patient record. Conclusions:: Our research shows that customization is feasible
+  in a medical digital library.'
+tags:
+- Clinical information system
+- Medical digital library
+- Multi-document information extraction
+- Multi-document summarization
+- User modeling
+links:
+- name: URL
+  url: https://doi.org/10.1016/j.artmed.2004.07.018
+---

content/publication/10-1016-j-ipm-2006-07-019/cite.bib

@@ -0,0 +1,19 @@
+@article{10.1016/j.ipm.2006.07.019,
+ abstract = {Web pages often embed scripts for a variety of purposes, including advertising and dynamic interaction. Understanding embedded scripts and their purpose can often help to interpret or provide crucial information about the web page. We have developed a functionality-based categorization of JavaScript, the most widely used web page scripting language. We then view understanding embedded scripts as a text categorization problem. We show how traditional information retrieval methods can be augmented with the features distilled from the domain knowledge of JavaScript and software analysis to improve classification performance. We perform experiments on the standard WT10G web page corpus, and show that our techniques eliminate over 50% of errors over a standard text classification baseline.},
+ address = {USA},
+ author = {Lu, Wei and Kan, Min-Yen},
+ doi = {10.1016/j.ipm.2006.07.019},
+ issn = {0306-4573},
+ issue_date = {March 2007},
+ journal = {Inf. Process. Manage.},
+ keywords = {ECMAScript, JavaScript, automated code classification, information retrieval, machine learning, program classification, program comprehension, program pattern, software metrics, source clone},
+ month = {March},
+ number = {2},
+ numpages = {14},
+ pages = {431–444},
+ publisher = {Pergamon Press, Inc.},
+ title = {Supervised categorization of JavaScriptTM using program analysis features},
+ url = {https://doi.org/10.1016/j.ipm.2006.07.019},
+ volume = {43},
+ year = {2007}
+}

content/publication/10-1016-j-ipm-2006-07-019/index.md

@@ -0,0 +1,36 @@
+---
+title: Supervised categorization of JavaScriptTM using program analysis features
+authors:
+- Wei Lu
+- Min-Yen Kan
+date: '2007-03-01'
+publishDate: '2024-10-20T00:58:41.583687Z'
+publication_types:
+- article-journal
+publication: '*Inf. Process. Manage.*'
+doi: 10.1016/j.ipm.2006.07.019
+abstract: Web pages often embed scripts for a variety of purposes, including advertising
+  and dynamic interaction. Understanding embedded scripts and their purpose can often
+  help to interpret or provide crucial information about the web page. We have developed
+  a functionality-based categorization of JavaScript, the most widely used web page
+  scripting language. We then view understanding embedded scripts as a text categorization
+  problem. We show how traditional information retrieval methods can be augmented
+  with the features distilled from the domain knowledge of JavaScript and software
+  analysis to improve classification performance. We perform experiments on the standard
+  WT10G web page corpus, and show that our techniques eliminate over 50% of errors
+  over a standard text classification baseline.
+tags:
+- ECMAScript
+- JavaScript
+- automated code classification
+- information retrieval
+- machine learning
+- program classification
+- program comprehension
+- program pattern
+- software metrics
+- source clone
+links:
+- name: URL
+  url: https://doi.org/10.1016/j.ipm.2006.07.019
+---

content/publication/10-1016-j-ipm-2007-03-010/cite.bib

@@ -0,0 +1,19 @@
+@article{10.1016/j.ipm.2007.03.010,
+ abstract = {We argue that the quality of a summary can be evaluated based on how many concepts in the original document(s) that can be preserved after summarization. Here, a concept refers to an abstract or concrete entity or its action often expressed by diverse terms in text. Summary generation can thus be considered as an optimization problem of selecting a set of sentences with minimal answer loss. In this paper, we propose a document concept lattice that indexes the hierarchy of local topics tied to a set of frequent concepts and the corresponding sentences containing these topics. The local topics will specify the promising sub-spaces related to the selected concepts and sentences. Based on this lattice, the summary is an optimized selection of a set of distinct and salient local topics that lead to maximal coverage of concepts with the given number of sentences. Our summarizer based on the concept lattice has demonstrated competitive performance in Document Understanding Conference 2005 and 2006 evaluations as well as follow-on tests.},
+ address = {USA},
+ author = {Ye, Shiren and Chua, Tat-Seng and Kan, Min-Yen and Qiu, Long},
+ doi = {10.1016/j.ipm.2007.03.010},
+ issn = {0306-4573},
+ issue_date = {November, 2007},
+ journal = {Inf. Process. Manage.},
+ keywords = {Text summarization, Semantic, Document concept lattice, Concept},
+ month = {November},
+ number = {6},
+ numpages = {20},
+ pages = {1643–1662},
+ publisher = {Pergamon Press, Inc.},
+ title = {Document concept lattice for text understanding and summarization},
+ url = {https://doi.org/10.1016/j.ipm.2007.03.010},
+ volume = {43},
+ year = {2007}
+}

content/publication/10-1016-j-ipm-2007-03-010/index.md

@@ -0,0 +1,35 @@
+---
+title: Document concept lattice for text understanding and summarization
+authors:
+- Shiren Ye
+- Tat-Seng Chua
+- Min-Yen Kan
+- Long Qiu
+date: '2007-11-01'
+publishDate: '2024-10-20T00:58:41.548792Z'
+publication_types:
+- article-journal
+publication: '*Inf. Process. Manage.*'
+doi: 10.1016/j.ipm.2007.03.010
+abstract: We argue that the quality of a summary can be evaluated based on how many
+  concepts in the original document(s) that can be preserved after summarization.
+  Here, a concept refers to an abstract or concrete entity or its action often expressed
+  by diverse terms in text. Summary generation can thus be considered as an optimization
+  problem of selecting a set of sentences with minimal answer loss. In this paper,
+  we propose a document concept lattice that indexes the hierarchy of local topics
+  tied to a set of frequent concepts and the corresponding sentences containing these
+  topics. The local topics will specify the promising sub-spaces related to the selected
+  concepts and sentences. Based on this lattice, the summary is an optimized selection
+  of a set of distinct and salient local topics that lead to maximal coverage of concepts
+  with the given number of sentences. Our summarizer based on the concept lattice
+  has demonstrated competitive performance in Document Understanding Conference 2005
+  and 2006 evaluations as well as follow-on tests.
+tags:
+- Text summarization
+- Semantic
+- Document concept lattice
+- Concept
+links:
+- name: URL
+  url: https://doi.org/10.1016/j.ipm.2007.03.010
+---

content/publication/10-1109-tasl-2007-911559/cite.bib

@@ -0,0 +1,18 @@
+@article{10.1109/TASL.2007.911559,
+ abstract = {We present LyricAlly, a prototype that automatically aligns acoustic musical signals with their corresponding textual lyrics, in a manner similar to manually-aligned karaoke. We tackle this problem based on a multimodal approach, using an appropriate pairing of audio and text processing to create the resulting prototype. LyricAlly's acoustic signal processing uses standard audio features but constrained and informed by the musical nature of the signal. The resulting detected hierarchical rhythm structure is utilized in singing voice detection and chorus detection to produce results of higher accuracy and lower computational costs than their respective baselines. Text processing is employed to approximate the length of the sung passages from the lyrics. Results show an average error of less than one bar for per-line alignment of the lyrics on a test bed of 20 songs (sampled from CD audio and carefully selected for variety). We perform a comprehensive set of system-wide and per-component tests and discuss their results. We conclude by outlining steps for further development.},
+ author = {Kan, Min-Yen and Wang, Ye and Iskandar, D. and New, Tin Lay and Shenoy, A.},
+ doi = {10.1109/TASL.2007.911559},
+ issn = {1558-7916},
+ issue_date = {February 2008},
+ journal = {Trans. Audio, Speech and Lang. Proc.},
+ keywords = {Acoustic signal detection, acoustic signal processing, music, text processing},
+ month = {February},
+ number = {2},
+ numpages = {12},
+ pages = {338–349},
+ publisher = {IEEE Press},
+ title = {LyricAlly: Automatic Synchronization of Textual Lyrics to Acoustic Music Signals},
+ url = {https://doi.org/10.1109/TASL.2007.911559},
+ volume = {16},
+ year = {2008}
+}

content/publication/10-1109-tasl-2007-911559/index.md

@@ -0,0 +1,36 @@
+---
+title: 'LyricAlly: Automatic Synchronization of Textual Lyrics to Acoustic Music Signals'
+authors:
+- Min-Yen Kan
+- Ye Wang
+- D. Iskandar
+- Tin Lay New
+- A. Shenoy
+date: '2008-02-01'
+publishDate: '2024-10-20T00:58:41.527763Z'
+publication_types:
+- article-journal
+publication: '*Trans. Audio, Speech and Lang. Proc.*'
+doi: 10.1109/TASL.2007.911559
+abstract: We present LyricAlly, a prototype that automatically aligns acoustic musical
+  signals with their corresponding textual lyrics, in a manner similar to manually-aligned
+  karaoke. We tackle this problem based on a multimodal approach, using an appropriate
+  pairing of audio and text processing to create the resulting prototype. LyricAlly's
+  acoustic signal processing uses standard audio features but constrained and informed
+  by the musical nature of the signal. The resulting detected hierarchical rhythm
+  structure is utilized in singing voice detection and chorus detection to produce
+  results of higher accuracy and lower computational costs than their respective baselines.
+  Text processing is employed to approximate the length of the sung passages from
+  the lyrics. Results show an average error of less than one bar for per-line alignment
+  of the lyrics on a test bed of 20 songs (sampled from CD audio and carefully selected
+  for variety). We perform a comprehensive set of system-wide and per-component tests
+  and discuss their results. We conclude by outlining steps for further development.
+tags:
+- Acoustic signal detection
+- acoustic signal processing
+- music
+- text processing
+links:
+- name: URL
+  url: https://doi.org/10.1109/TASL.2007.911559
+---