The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Language Model (LM) constitutes one of the key components in Keyword Spotting (KWS). The rapid development of the World Wide Web (WWW) makes it an extremely large and valuable data source for LM training, but it is not optimal to use the raw transcripts from WWW due to the mismatch of content between the web corpus
This paper presents a text query-based method for keyword spotting from online Chinese handwritten documents. The similarity between a text word and handwriting is obtained by combining the character similiarity scores given by a character classifier. To overcome the ambiguity of character segmentation, multiple
topic analysis of LDA for feature selection and compare it with the classical feature selection metrics in text categorization. For the experiments, we use SVM as the classifier and tf*idf weighting for weighting the terms. We observed that almost in all metrics, information gain performs best at all keyword numbers while
Spotting keywords in handwritten documents without transcription is a valuable method as it allows one to search, index, and classify such documents. In this paper we show that keyword spotting based on bi-directional Long Short-Term Memory (BLSTM) recurrent neural nets can successfully be applied on online
This paper presents a corpus-based approach for extracting keywords from a text written in a language that has no word boundary. Based on the concept of Thai character cluster, a Thai running text is preliminarily segmented into a sequence of inseparable units, called TCCs. To enable the handling of a large-scaled
Keyword extraction has been a very traditional topic in Natural Language Processing. However, most methods have been too complicated and slow to be applied in real applications, for example in web-based system. This paper proposes an approach which will complete some preparing works focusing on exploring the
Being able to search for words or phrases in historic handwritten documents is of paramount importance when preserving cultural heritage. Storing scanned pages of written text can save the information from degradation, but it does not make the textual information readily available. Automatic keyword spotting systems
To reduce the human effort in labeling the training set for document classification, some learning algorithms ask users to give the representative keywords for each class rather than any labeled documents. The key challenge in such \emph {keyword-labeled classification} is how to learn the high quality classifier with
Keywords are the critical resources of information management and retrieval, automatic text classification and clustering. The keywords extraction plays an important role in the process of constructing structured text. Current algorithms of keywords extraction have matured in some ways. However the errors of word
metrics used in text categorization by using local and global policies. For the experiments, we use three datasets which vary in size, complexity and skewness. We use SVM as the classifier and tf-idf weighting for term weighting. We observed that almost in all metrics, local policy outperforms when the number of keywords is
This work proposes an approach to address the problem of improving content selection in automatic text summarization by using probabilistic neural network (PNN). This approach is a trainable summarizer, which takes into account several features, including sentence position, positive keyword, negative keyword, sentence
Due to the exponential growth of available text documents in digital form, it is of great importance to develop techniques for automatic document classification based on the textual contents. Earlier document classification techniques have used keyword-based features and related statistics to achieve good results when
huge irrelevant search hits. In this paper, we propose an improved method for ranking of search results to reduce human efforts on locating interesting hits. The search results are re-ranked using adaptive user interest hierarchies (AUIH), which considers both investigator-defined keywords and user interest learnt from
Automatically assigning relevant text keywords to images is an important problem. Many algorithms have been proposed in the past decade and achieved good performance. Efforts have focused upon model representations of keywords, but properties of features have not been well investigated. In most cases, a group of
Keywords normally carry large amount of category information. In order to fully utilize this kind of information for text classification, this paper proposes a new text feature conversion method based on the SKG model. The method uses the classified texts with the listed key words as the training data to train the
The Fisher kernel is a generic framework which combines the benefits of generative and discriminative approaches to pattern classification. In this contribution, we propose to apply this framework to handwritten word-spotting. Given a word image and a keyword generative model, the idea is to generate a vector which
. First, the related textual information associated with Web images is identified as the candidate annotations for Web images. Second, the word co-occurrence is utilized to eliminate irrelevant keywords for improving the annotation accuracy. Then, the keyword-based association analysis is exploited to further discover
results in up to 1.1% absolute Word Error Rate (WER) improvement as compared to keyword-based approaches. The proposed approach reduces the WER by 6.3% absolute in our experiments, compared to an in-domain LM without considering any Web data.
mechanisms with a traditional indexing method. The goal is to identify a higher semantic content and more meaningful keyword combinations, considering both supervised and unsupervised techniques. Within a specific implementation both Bayesian learning as well as clustering are integrated to support a boost parameter towards
of cultural information. Therefore, text categorization research has become more important. The paper improved the precision of the traditional text categorization by the process that we mended the weight of words and mined potential keywords, then found their relationship. In the end of the paper, an experiment was
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.