This paper proposes a novel representation for Authorship Attribution (AA), based on Concise Semantic Analysis (CSA), which has been successfully used in Text Categorization (TC). Our approach for AA, called Document Author Representation (DAR), builds document vectors in a space of authors, calculating the relationship between textual features and authors. In order to evaluate our approach, we...

In this paper we present a new, abstract model for textual data objects with embedded markup. Based on the model, we propose a uniform representation for these objects that borrows its concrete syntax from the ISO standard SGML. Such a uniform representation will greatly facilitate the development of software that analyzes, formats or otherwise processes these objects. We then describe a toolse...

With the great and rapidly growing number of documents available in digital form (Internet, library, CD-Rom...), the automatic classification of texts has become a significant research field and a fundamental task in document processing. This paper deals with unsupervised classification of textual documents also called text clustering using Self-Organizing Maps of Kohonen in two new situations:...

Concept mapsTM have existed in the educational community for some time. A concept map is a visual representation of a person’s (student’s) knowledge of a domain. Many have reported on computer-based implementations of interactive concept map building tools. However, existing concept webs are rooted in a propositional, primarily textual, knowledge representation scheme. Further, existing compute...

我們所參與公開評測 NTCIR10 RITE-2[5]將文字蘊涵的研究分成兩種層面,首先是分兩 類(Binary Class, BC) ,任務的目標是單純判別 T1 與 T2 之間是否具有蘊涵關係。但句 子之間蘊涵關係並不能單純以有或沒有這麼簡單就區分開,NTCIR RITE 另外定義多類 (Multi Class, MC)這項任務,將句子之間的蘊涵分類為正向、雙向、矛盾、與獨立四種 關係。假設這個句子對具有蘊涵關係,但有可能兩個句子所包涵的資訊數量不同,造成 我們只能從其中一個句子推論出另一個句子的完整的意思,這樣的情況我們稱為兩個句 子間的蘊涵關係為正向蘊涵。反之兩個句子可以互相推論出另一個句子的含意,這樣的 情況我們就稱為雙向蘊涵關係。假設句子對之間沒有蘊涵關係,我們可以很合理認為兩 個句子所表達的意思不相同,但這並不完全正確的想法。可能兩個句子所包涵的資訊大 致相同只是少部份...

This paper describes the Recognizing Textual Entailment (RTE) system that our teams developed for TAC 2011. Our system combines the entailment score calculated by lexicallevel matching with the machine-learningbased filtering mechanism using various features obtained from lexical-level, chunk-level and predicate argument structure-level information. In the filtering mechanism, we try to discard...

The paper introduces a framework for representation and acquisition of knowledge emerging from large samples of textual data. We utilise a tensor-based, dis-tributional representation of simple statements extracted from text, and show how one can use the representation to infer emergent knowledge patterns from the tex-tual data in an unsupervised manner. Examples of the patterns we investigate ...

Unsupervised learning of low-dimensional, semantic representations of words and entities has recently gained aŠention. In this paper we describe the Semantic Entity Retrieval Toolkit (SERT) that provides implementations of our previously published entity representation models. Œe toolkit provides a uni€ed interface to di‚erent representation learning algorithms, €ne-grained parsing con€guration...

Graph-based formalisms provide an intuitive and easily understandable vehicle for knowledge representation. In this paper several existing graph-based formalisms are described. Furthermore, a new graph-based formalism for knowledge representation is defined. Basic concepts for graphical representation (nodes and links) as well as their variation are described. Context node, context link and pro...