1,721,140 research outputs found
Joint Modeling of Topics, Citations, and Topical Authority in Academic Corpora
No full textMuch of scientific progress stems from previously published findings, but searching through the vast sea of scientific publications is difficult. We often rely on metrics of scholarly authority to find the prominent authors but these authority indices do not differentiate authority based on research topics. We present Latent Topical-Authority Indexing (LTAI) for jointly modeling the topics, citations, and topical authority in a corpus of academic papers. Compared to previous models, LTAI differs in two main aspects. First, it explicitly models the generative process of the citations, rather than treating the citations as given. Second, it models each author’s influence on citations of a paper based on the topics of the cited papers, as well as the citing papers. We fit LTAI into four academic corpora: CORA, Arxiv Physics, PNAS, and Citeseer. We compare the performance of LTAI against various baselines, starting with the latent Dirichlet allocation, to the more advanced models including author-link topic model and dynamic author citation topic model. The results show that LTAI achieves improved accuracy over other similar models when predicting words, citations and authors of publications.11
Exact Truthmaker Semantics for Modal Logics
No full textThe present paper attempts to provide an exact truthmaker semantical analysis of modalized propositions. According to the present proposal, an exact truthmaker for "Necessarily P" is a state that bans every exact truthmaker for "Not P", and an exact truthmaker for "Possibly P" is a state that allows an exact truthmaker for P. Based on this proposal, a formal semantics will be developed; and the soundness and completeness results for a well-known family of the systems of normal modal propositional logic will be established. It shall be seen that the present analysis offers an exactification of the standard Kripke semantics in the sense that it analyzes the accessibility relation between possible worlds in terms of the banning and allowing relations between the constituent states, and thereby gives an account of "truth at a possible world" in terms of exact truthmaking.
The Kripkean explanation of aposteriori necessity: in the case of identity statements about chemical substances
No full textIn the addenda to his Naming and Necessity, Kripke provides an account of how necessary aposteriori statements are possible. In such a case, there is an apriori general principle telling us that it is necessary if true at all. Though straightforward in its broad compass, this account faces two obvious questions in its application: in each case of necessary aposteriori statements, what is the underlying principle and how is it established apriori? I treat these questions with respect to theoretical identity statements concerning chemical substances, such as 'water is H2O'. I argue that the general principle underlying the necessity of the statements is that if a chemical substance has a certain chemical composition, then it could not have had any other chemical composition. Then I defend the view that the principle is a conceptual truth by providing a novel derivation of it from the theoretical concept of chemical substance with a sufficient level of formal rigor. The logical principles required for the derivation will also be stated and defended.
Analysing user identity via time-sensitive semantic edit distance (t-SED): a case study of Russian trolls on Twitter
No full textIn the digital era, individuals are increasingly profiled and grouped based on the traces that they leave behind in online social networks such as Twitter and Facebook. In this paper, we develop and evaluate a novel text analysis approach for studying user identity and social roles by redefining identity as a sequence of timestamped items (e.g., tweet texts). We operationalise this idea by developing a novel text distance metric, the time-sensitive semantic edit distance (t-SED), which accounts for the temporal context across multiple traces. To evaluate this method, we undertake a case study of Russian online-troll activity within US political discourse. The novel metric allows us to classify the social roles of trolls based on their traces, in this case tweets, into one of the predefined categories left-leaning, right-leaning, and news feed. We show the effectiveness of the t-SED metric to measure the similarities between tweets while accounting for the temporal context, and we use novel data visualisation techniques and qualitative analysis to uncover new empirical insights into Russian troll activity that have not been identified in the previous work. In addition, we highlight a connection with the field of actor–network theory and the related hypotheses of Gabriel Tarde, and we discuss how social sequence analysis using t-SED may provide new avenues for tackling a longstanding problem in social theory: how to analyse society without separating reality into micro vs. macro-levels.11
Necessity, Essence, and Explanation
No full textI shall discuss some of the relations among metaphysical modality, essence, and explanation. Marion Godman, Antonella Mallozzi and David Papineau have recently argued that the essence of a kind consists in its super-explanatory property-a single property that is causally responsible for a multitude of commonalities shared by the instances of the kind. And they argue that this super-explanatory account of essence offers a principled account of aposteriori necessities concerning kinds. I shall examine their arguments and argue that they are fallacious. Along the way, a general problem will also emerge that applies to any account that tries to explicate the notion of essence in terms of an explanatory relation.
Hierarchical Dirichlet scaling process
Full text linkWe present the hierarchical Dirichlet scaling process (HDSP), a Bayesian nonparametric mixed membership model. The HDSP generalizes the hierarchical Dirichlet process to model the correlation structure between metadata in the corpus and mixture components. We construct the HDSP based on the normalized gamma representation of the Dirichlet process, and this construction allows incorporating a scaling function that controls the membership probabilities of the mixture components. We develop two scaling methods to demonstrate that different modeling assumptions can be expressed in the HDSP. We also derive the corresponding approximate posterior inference algorithms using variational Bayes. Through experiments on datasets of newswire, medical journal articles, conference proceedings, and product reviews, we show that the HDSP results in a better predictive performance than labeled LDA, partially labeled LDA, and author topic model and a better negative review classification performance than the supervised topic model and SVM.11Nsciescopu
- …
