SenU-PTM: a novel phrase-based topic model for short-text topic discovery by exploiting word embeddings

PurposeTopic model has been widely applied to discover important information from a vast amount of unstructured data. Traditional long-text topic models such as Latent Dirichlet Allocation may suffer from the sparsity problem when dealing with short texts, which mostly come from the Web. These models also exist the readability problem when displaying the discovered topics. The purpose of this paper is to propose a novel model called the Sense Unit based Phrase Topic Model (SenU-PTM) for both the sparsity and readability problems.Design/methodology/approachSenU-PTM is a novel phrase-based short-text topic model under a two-phase framework. The first phase introduces a phrase-generation algorithm by exploiting word embeddings, which aims to generate phrases with the original corpus. The second phase introduces a new concept of sense unit, which consists of a set of semantically similar tokens for modeling topics with token vectors generated in the first phase. Finally, SenU-PTM infers topics based on the above two phases.FindingsExperimental results on two real-world and publicly available datasets show the effectiveness of SenU-PTM from the perspectives of topical quality and document characterization. It reveals that modeling topics on sense units can solve the sparsity of short texts and improve the readability of topics at the same time.Originality/valueThe originality of SenU-PTM lies in the new procedure of modeling topics on the proposed sense units with word embeddings for short-text topic discovery.

Advanced T-LSIM System Detections using Amplified External Isolated Source-Sense Unit

Thermal-Laser Signal Injection Microscopy (T-LSIM) is a widely used fault isolation technique. Although there are several T-LSIM systems on the market, each is limited in terms of the voltage and current it can produce. In this paper, the authors explain how they incorporated an Amplified External Isolated Source-Sense (AxISS) unit into their T-LSIM platform, increasing its current sourcing capability and voltage biasing range. They also provide examples highlighting the types of faults and failures that the modified system can detect.

The Late Medieval Greek Vernacular Πολιτικὸς Στίχος Poetry: A Modern Linguistic Analysis into Intonation Units

The Late Medieval Greek “vernacular” (12th–15th c.) is one of the least studied stages of the history of the Greek language. The lack of interest by linguists can presumably be ascribed to its major source, i.e. metrical πολιτικὸς στίχος poetry. The language of this type of poetry has been labelled a “Kunstsprache”, because of its oral-formulaic character and because of its mixed idiom incorporating vernacular yet also archaizing elements. In this article, however, I demonstrate that the Late Medieval Greek πολιτικὸς στίχος poetry should not automatically be excluded from linguistic research, given that it clearly possesses a strongly vernacular, i.e. spoken, syntactic base: its underlying syntax runs in a very natural way. This is proven by the fact that we can apply the modern linguistic concept of the Intonation Unit, the basic unit of analysis in contemporary spoken(!) languages, to the texts composed in the πολιτικὸς στίχος: far from having an artificial syntax, the πολιτικὸς στίχος poetry is conceptually made up of short, simple “chunks” of information. More precisely, each verse consists of two (stylized) Intonation Units, demarcated by the fixed caesura, which can thus be equated with an Intonation Unit boundary. This thesis is supported by various arguments, both of a metrical and of a syntactico-semantic nature. Arguments belonging to the former category are the length of each half-line, the possibility of stress on the first syllable of each half-line, the origin of the metre, and especially the avoidance of elision at the caesura. In the second category of (syntactico-semantic) evidence, we can consider the tendency of each half-line for constituting a grammatical sense-unit. I also bring forward some little-studied syntactic features of Late Medieval Greek: first, I pay attention to the distribution of the archaizing “Wackernagel particles”, which do not only appear in second position in the verse, but also occur after the first word/constituent following the caesura and thus further confirm my thesis. The same holds for the position of “corrective afterthoughts”, for the verbs and pronominal objects taking the singular are consistently separated from their plural referents by the caesura. Once the Intonation Unit is thus established as a meaningful methodological tool for the analysis of the πολιτικὸς στίχος poetry, the way is cleared for more linguistic research on the Late Medieval Greek vernacular.

Users' Fuzzy Cognition Knowledge Learning in Interactive Evolutionary Computation and its Application

user in interactive evolutionary computation (IEC) has the characteristic of fuzzy cognition. Based on this, a method to learn users’ fuzzy cognition knowledge is given. The method includes the fuzzy expression of the basic elements of IEC such as search space, population, gene sense unit and so on. Then a method to increase the performance of IEC based on the knowledge of users’ fuzzy cognition is given. The above results enrich the researches of IEC users' cognition.

Constraints on intonational phrasing in English

This paper argues that intonational phrasing in English is subject to two constraints formulated purely in terms of surface syntax. The first applies to all headed structures, and the second to all structures of coordination. It is claimed that these constraints account for the facts more adequately than Selkirk's Sense Unit Condition. Several problematic types of constructions are discussed in this context, including structures with parentheticals. In addition to the syntax, accent placement is also shown to be relevant to intonational phrasing. Finally, proposals are made for the incorporation of the syntactic constraints in an HPSG grammar of English.