JRS’2012 Data Mining Competition: Topical Classification of Biomedical Research Papers

Janusz, Andrzej; Nguyen, Hung Son; Ślęzak, Dominik; Stawicki, Sebastian; Krasuski, Adam

doi:10.1007/978-3-642-32115-3_50

Cited by 11 publications

(10 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As the quality measures we use average values of the F 1 -score, P recision and Recall, obtained for all the test documents by comparing the predicted tags to those which were assigned by experts or authors. This type of evaluation of a tagging quality is popular for the multi-label classification problems [15].…”

Section: Methodsmentioning

confidence: 99%

Adaptive Learning for Improving Semantic Tagging of Scientific Articles

Janusz

Stawicki

Nguyen

2014

Annals of Computer Science and Information Systems

Self Cite

View full text Add to dashboard Cite

Abstract-In this paper we consider a problem of automatic labeling of textual data with concepts explicitly defined in an external knowledge base. We describe our tagging system and we also present a framework for adaptive learning of associations between terms or phrases from the texts and the concepts. Those associations are then utilized by our semantic interpreter, which is based on the Explicit Semantic Analysis (ESA) method, in order to label scientific articles indexed by our SONCA platform. Apart from the description of the learning algorithm, we show a few practical application examples of our system, in which it was used for tagging scientific articles with headings from the MeSH ontology, categories from ACM Computing Classification System and from OECD Fields of Science and Technology Classification.

show abstract

Section: Methodsmentioning

confidence: 99%

Adaptive Learning for Improving Semantic Tagging of Scientific Articles

Janusz

Stawicki

Nguyen

2014

Annals of Computer Science and Information Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…It can be generalized even further by, for example, considering the exact similarity function values during the voting or assigning weights to training objects that express their representativeness for the decision class. The k-nearest neighbours algorithm can also be used to predict values of a numeric decision attribute (regression) or to perform a multi-label classification [32]. However, in all those applications the correct selection of a similarity model is the factor that has the biggest influence on the quality of predictions.…”

Section: Similarity In Machine Learningmentioning

confidence: 98%

“…There are also several other research directions of the author that had a significant influence on the design of the proposed similarity learning models. Among them, the most important considered the problem of feature selection and learning with ensembles of single and multi-label classifiers [30][31][32][33][34][35][36][37]. Moreover, the research on unsupervised version of Rule-Based Similarity was largely influenced by the author's previous work on the semantic information retrieval and Explicit Semantic Analysis, which was conducted within the SYNAT project [38][39][40].…”

Section: Main Contributionsmentioning

confidence: 99%

“…Therefore, the approximation of concepts (a classification problem) can sometimes be restated as a prediction of decision attribute values for objects which are not included in the training set. In many practical applications, such as the topical classification of textual documents [32,35], it might be convenient to define more than one decision attribute. In such a case, a decision system will be denoted by…”

Section: Introduction To Rough Setsmentioning

confidence: 99%

See 1 more Smart Citation

Algorithms for Similarity Relation Learning from High Dimensional Data

Janusz

2014

Transactions on Rough Sets XVII

Self Cite

View full text Add to dashboard Cite

Abstract. The notion of similarity plays an important role in machine learning and artificial intelligence. It is widely used in tasks related to a supervised classification, clustering, an outlier detection and planning. Moreover, in domains such as information retrieval or case-based reasoning, the concept of similarity is essential as it is used at every phase of the reasoning cycle. The similarity itself, however, is a very complex concept that slips out from formal definitions. A similarity of two objects can be different depending on a considered context. In many practical situations it is difficult even to evaluate the quality of similarity assessments without considering the task for which they were performed. Due to this fact the similarity should be learnt from data, specifically for the task at hand. This paper presents a research on the problem of similarity learning, which is a part of author's PHD dissertation. It describes a similarity model, called Rule-Based Similarity, and shows algorithms for constructing this model from available data. The model utilizes notions from the rough set theory to derive a similarity function that allows to approximate the similarity relation in a given context. It is largely inspired by the idea of Tversky's feature contrast model and it has several analogical properties. In the paper, those theoretical properties are described and discussed. Moreover, the paper presents results of experiments on real-life data sets, in which a quality of the proposed model is thoroughly evaluated and compared with the state-of-the-art algorithms.

show abstract

“…A form of this challenge was similar to the competitions which our team organized in the past [7], [8] on the TunedIT platform [9]. This time, however, we organized it on a novel web platform called Knowledge Pit (www.challenge.mimuw.edu.pl) hoping that the participants will be able to enrich our understanding of the EWID data and point at attributes that describe the most relevant information to the stated problem.…”

mentioning

confidence: 99%

Key Risk Factors for Polish State Fire Service: a Data Mining Competition at Knowledge Pit

Janusz¹,

Krasuski²,

Stawicki³

et al. 2014

Annals of Computer Science and Information Systems

Self Cite

View full text Add to dashboard Cite

Abstract-In this paper we summarize AAIA'14 Data Mining Competition: Key risk factors for Polish State Fire Service which was held between February 3, 2014 and May 5, 2014 at the Knowledge Pit platform http://challenge.mimuw.edu.pl/. We describe the scope and background of this competition and we explain in details the evaluation procedure. We also briefly overview the results of this analytical challenge, showing the way in which those results can be beneficial to one of our other projects which is related to the problem of improving firefighter safety at a fire scene. Finally, we reveal some technical details regarding the architecture and functionalities of the Knowledge Pit competition platform, which we are developing in order to facilitate solving of practical problems that require advanced data analytics.

show abstract

JRS’2012 Data Mining Competition: Topical Classification of Biomedical Research Papers

Cited by 11 publications

References 6 publications

Adaptive Learning for Improving Semantic Tagging of Scientific Articles

Adaptive Learning for Improving Semantic Tagging of Scientific Articles

Algorithms for Similarity Relation Learning from High Dimensional Data

Key Risk Factors for Polish State Fire Service: a Data Mining Competition at Knowledge Pit

Contact Info

Product

Resources

About