Feature selection strategies for automated classification of digital media content

Cobo, Ángel

doi:10.1177/0165551511412028

Cited by 5 publications

(8 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most of the text document categorisation is adopted for the vector space model [27–30] to represent the documents, that is to say, each unique term in vocabulary represents one dimension in the future vector space. Therefore, the text document data sets can be represented as a document-by-term matrix D ( n × m ) , where n and m indicate the number of documents and the number of terms occurring in the document data set, respectively.…”

Section: Preliminariesmentioning

confidence: 99%

DIC-DOC-K-means: Dissimilarity-based Initial Centroid selection for DOCument clustering using K-means for improving the effectiveness of text document clustering

Lakshmi

Baskar²

2018

Journal of Information Science

View full text Add to dashboard Cite

In this article, a new initial centroid selection for a K-means document clustering algorithm, namely, Dissimilarity-based Initial Centroid selection for DOCument clustering using K-means (DIC-DOC- K-means), to improve the performance of text document clustering is proposed. The first centroid is the document having the minimum standard deviation of its term frequency. Each of the other subsequent centroids is selected based on the dissimilarities of the previously selected centroids. For comparing the performance of the proposed DIC-DOC- K-means algorithm, the results of the K-means, K-means++ and weighted average of terms-based initial centroid selection + K-means (Weight_Avg_Initials + K-means) clustering algorithms are considered. The results show that the proposed DIC-DOC- K-means algorithm performs significantly better than the K-means, K-means++ and Weight_Avg_Initials+ K-means clustering algorithms for Reuters-21578 and WebKB with respect to purity, entropy and F-measure for most of the cluster sizes. The cluster sizes used for Reuters-8 are 8, 16, 24 and 32 and those for WebKB are 4, 8, 12 and 16. The results of the proposed DIC-DOC- K-means give a better performance for the number of clusters that are equal to the number of classes in the data set.

show abstract

Section: Preliminariesmentioning

confidence: 99%

DIC-DOC-K-means: Dissimilarity-based Initial Centroid selection for DOCument clustering using K-means for improving the effectiveness of text document clustering

Lakshmi

Baskar²

2018

Journal of Information Science

View full text Add to dashboard Cite

show abstract

“…Three major German and Austrian television channels have published their guidelines for indexing (ARD/ORF/ZDF, ) the content of programs. Research, however, is mainly focused on automatic classification and text mining of news items (e.g., Rocha & Cobo, ) rather than conceptual indexing. Indeed, news material from early on constituted an important part of information retrieval (IR) test collections, such as the TREC collections, the first example being Salton's Time magazine collection (Sanderson, ).…”

Section: Previous Workmentioning

confidence: 99%

Decentralized subject indexing of television programs: The effects of using a semicontrolled indexing language

Søbak

Pharo

2016

Asso for Info Science & Tech

View full text Add to dashboard Cite

We performed an exploratory case study to understand how subject indexing performed by television production staff using a semicontrolled vocabulary affects indexing quality. In the study we used triangulation, combining tag analysis and semistructured interviews, with production staff of the Norwegian Broadcasting Corporation. The main findings reveal incomplete indexing of TV programs and their parts, in addition to low indexing consistency and uneven indexing exhaustivity. The informants expressed low motivation and a high level of uncertainty regarding the task. Internal guidelines and high domain knowledge among the indexers does not form a sufficient basis for creating quality and consistency in the vocabulary. The challenges that are revealed in the terminological analysis, combined with low indexing knowledge and lack of motivation, will create difficulties in the retrieval phase.

show abstract

“…Furthermore, there has been mounting interest in organizing news and newspaper articles using automatic techniques such as text classification and indexing algorithms (e.g. Chen and Lin 2000;Evans and Klavans 2003;Casillas et al 2003;Mamakis et al 2011;Rocha and Cobo 2011), or, on occasion, those based on user-features, in a usercentered fashion, such as the automatic summarization and categorization of news derived from user choices (Banos et al 2006), user modeling (Wongchokprasitti and Brusilovsky 2007), or user profiles (Bouras and Tsogkas 2010). However, as some authors have pointed out, automatic indexing and user-based retrieval systems such as Google's are not exempt from bias or subjectivity either (e.g.…”

Section: Media Knowledge Organization In the Context Of Ko Studiesmentioning

confidence: 99%

Critical Organization of Knowledge in Mass Media Information Systems

García-Gutiérrez¹,

Martínez-Ávila²

2014

View full text Add to dashboard Cite

Antonio García Gutiérrez holds a Ph.D. in information sciences from Complutense University of Madrid and is a professor at the School of Communication in Seville University. He has acted as an information systems consultant for international organizations such as UNESCO, World Tourism Organization and as an evaluator of scientific projects for the Fifth Framework Programme of the European Union. He has written more than fifteen books that address the problems of knowledge organization and its construction from post-epistemological stances linked to objects related to cultural studies and the postcolonial approach, such as the memory, the experience, the identity and the culture. Abstract: This paper studies knowledge organization (KO) in media archives, focusing on the presence of subjectivity in the core tasks of mass media knowledge organizers (MKOS) dealing with press, radio and TV records, such as classification, representation, and any other process related to content analysis and organization in news information systems. Far from rejecting subjectivity and ideological bias in these operations -since they coparticipate in the media construction of reality-the authors consider MKOS to be genuine ideological and cultural mediators with the right and social responsibility to explicitly state the results of their "objectifiable" work (obtained through KO protocols and procedures determined by the media/company, classifications, thesauri, ontologies, etc.) and differentiate them from those of their political, ideological, cultural and, in sum, subjective stances. In order to achieve this, we propose the application of critical operators that should be followed by technical, collaborative and even technological actions geared to investing information systems with the capacity to consider those stances and allowing users to distinguish them. In short, it is the theoretical recognition of the subjective and biased presence of media knowledge organization operators in a job that is usually considered neutral, banal and even objective, and the initial development of tools for critical, self-critical, technical, and technological training keyed to its practical solution. This paper outlines the lines of work of a broader research study on the critical function of KO in the field of global media memory.

show abstract

Feature selection strategies for automated classification of digital media content

Cited by 5 publications

References 17 publications

DIC-DOC-K-means: Dissimilarity-based Initial Centroid selection for DOCument clustering using K-means for improving the effectiveness of text document clustering

DIC-DOC-K-means: Dissimilarity-based Initial Centroid selection for DOCument clustering using K-means for improving the effectiveness of text document clustering

Decentralized subject indexing of television programs: The effects of using a semicontrolled indexing language

Critical Organization of Knowledge in Mass Media Information Systems

Contact Info

Product

Resources

About