Long gram-based indices are experimented at NTCIR-3 patent task. To make gram-based indices, no analyses such as morphological ones are required. The docno, abj, clj and dej tag fields are extracted from NTCIR-3 patent corpus. The total index size is 11.4Gbyte and time to make indices is about 8.7 hours. Median search time per word from abj and dej index is 9.8msec and 91.8msec respectively. Av...

Folksonomies have shown interesting potential for improving information discovery and exploration. Recent folksonomy systems investigate on the use of tag assignments, which combine Web resources with annotations (tags), and the users that have created the annotations. In this paper, we investigate on the effect of grouping resources, i.e. creating sets of resources, and using this additional s...

The development of FrameNet, a large database of semantically annotated sentences, has primed research into statistical methods for semantic tagging. We advance previous work by adopting a Maximum Entropy approach and by using Viterbi search to find the highest probability tag sequence for a given sentence. Further we examine the use of syntactic pattern based re-ranking to further increase per...

Long gram-based indices are experimented at NTCIR-4 patent task. No morphological analyses are required to make gram-based indices. The ABJ and DEJ tag fields are extracted and indexed from NTCIR-4 patent corpus. Passages are extracted and indexed also. The total index size is 240Gbyte and time to make indices is about 86 hours. By merging the result of passage retrieval with the result of docu...

In this paper a new graph-based model is proposed for the representation of textual documents. Graph-structures are obtained from textual documents by making use of the well-known Part-OfSpeech (POS) tagging technique. More specifically, a simple rule-based (re)classifier is used to map each tag onto graph vertices and edges. As a result, a decomposition of textual documents is obtained where t...

We describe the two systems submitted by the DCU-Symantec team to Task 1.1. of the WMT 2013 Shared Task on Quality Estimation for Machine Translation. Task 1.1 involve estimating postediting effort for English-Spanish translation pairs in the news domain. The two systems use a wide variety of features, of which the most effective are the word-alignment, n-gram frequency, language model, POS-tag...

Social tagging systems, such as Delicious, My Web 2.0, Flickr, YouTube, have been very successful and attracted hundreds of million users. User provided tags of an object/page can be used to help the user re-find the object through search or share the customized object with other people. Instead of waiting for a user to find and input the appropriate words to tag an object, we propose to automa...

University rankings generally present users with the problem of placing the results given for an institution in context. Only a comparison with the performance of all other institutions makes it possible to say exactly where an institution stands. In order to interpret the results of the SCImago Institutions Ranking (based on Scopus data) and the Leiden Ranking (based on Web of Science data), i...

Part I Social Media Content Analysis Quantifying Visual-Representativeness of Social Image Tags Using Image Tag Clarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Aixin Sun and Sourav S. Bhowmick Tag-Based Social Image Search: Toward Relevant and Diverse Results . . 25 Kuiyuan Yang, Meng Wang, Xian-Sheng Hua, and Hong-Jiang Zhang Social Image Tag Ranking by Two-View Learning . . ...