Syntactic Conventions Research Articles

Automated description generation for software patches

Software patches are pivotal in refining and evolving codebases, addressing bugs, vulnerabilities, and optimizations. Patch descriptions provide detailed accounts of changes, aiding comprehension and collaboration among developers. However, manual description creation poses challenges in terms of time consumption and variations in quality and detail. In this paper, we propose PatchExplainer, an approach that addresses these challenges by framing patch description generation as a machine translation task. In PatchExplainer, we leverage explicit representations of critical elements, historical context, and syntactic conventions. Moreover, the translation model in PatchExplainer is designed with an awareness of description similarity. Particularly, the model is explicitly trained to recognize and incorporate similarities present in patch descriptions clustered into groups, improving its ability to generate accurate and consistent descriptions across similar patches. The dual objectives maximize similarity and accurately predict affiliating groups. Our experimental results on a large dataset of real-world software patches show that PatchExplainer consistently outperforms existing methods, with improvements up to 189% in BLEU, 5.7X in Exact Match rate, and 154% in Semantic Similarity, affirming its effectiveness in generating software patch descriptions.

Controlling the taxonomic variable: Taxonomic concept resolution for a southeastern United States herbarium portal

Overview. Taxonomic names are imperfect identifiers of specific and sometimes conflicting taxonomic perspectives in aggregated biodiversity data environments. The inherent ambiguities of names can be mitigated using syntactic and semantic conventions developed under the taxonomic concept approach. These include: (1) representation of taxonomic concept labels (TCLs: name sec. source) to precisely identify name usages and meanings, (2) use of parent/child relationships to assemble separate taxonomic perspectives, and (3) expert provision of Region Connection Calculus articulations (RCC–5: congruence, [inverse] inclusion, overlap, exclusion) that specify how data identified to different-sourced TCLs can be integrated. Application of these conventions greatly increases trust in biodiversity data networks, most of which promote unitary taxonomic 'syntheses' that obscure the actual diversity of expert-held views. Better design solutions allow users to control the taxonomic variable and thereby assess the robustness of their biological inferences under different perspectives. A unique constellation of prior efforts – including the powerful Symbiota collections software platform, the Euler/X multi-taxonomy alignment toolkit, and the "Weakley Flora" which entails 7,000 concepts and more than 75,000 RCC–5 articulations – provides the opportunity to build a first full-scale concept resolution service for SERNEC, the SouthEast Regional Network of Expertise and Collections, currently with 60 member herbaria and 2 million occurrence records. Intellectual merit. We have developed a multi-dimensional, step-wise plan to transition SERNEC's data culture from name- to concept-based practices. (1) We will engage SERNEC experts through annual, regional workshops and follow-up interactions that will foster buy-in and ultimately the completion of 12 community-identified use cases. (2). We will leverage RCC–5 data from the Weakley Flora and further development of the Euler/X logic reasoning toolkit to provide comprehensive genus- to variety-level concept alignments for at least 10 major flora treatments with highest relevance to SERNEC. The visualizations and estimated > 1 billion inferred concept-to-concept relations will effectively drive specimen data integration in the transformed portal. (3) We will expand Symbiota's taxonomy and occurrence schemas and related user interfaces to support the new concept data, including novel batch and map-based specimen determination modules, with easy output options in Darwin Core Archive format. (4) Through combinations of the new technology, enlisted taxonomic expertise, and SERNEC's large image resources, we will upgrade minimally 80% of all SERNEC specimen identifications from names to the narrowest suitable TCLs, or add "uncertainty" flags to specimens needing further study. (5) We will utilize the novel tools and data to demonstrate how controlling for the taxonomic variable in 12 use cases variously drives the outcomes of evolutionary, ecological, and conservation-based research hypotheses. Broader impacts. Our project is focused on just one herbarium network, but the potential impact is as wide as Darwin Core or even comparative biology. We believe that trust in networked biodiversity data depends on open and dynamic system designs, allowing expert access and resolution of multiple conflicting views that reflect the complex realities of ongoing taxonomic research. Taking well over 1 million SERNEC records from name- to TCL-resolution will show that "big" specimen data can pass the credibility threshold needed to validate the substantive data mobilization investment. We will mentor one postdoctoral researcher (UNC), two Ph.D. students (ASU, UIUC), and at least 15 undergraduate students (ASU). Each of our workshops will capacitate 10-15 SERNEC experts, who in turn can recruit colleagues and students at their home collections. We will incorporate the project theme and use cases into undergraduate courses taught at six institutions and reaching an estimated 300-500 students annually (10-40% minority students). At each institution, project members will make a systematic effort to recruit new students from underrepresented groups. Our group's leadership of Symbiota (with close ties to iDigBio), SERNEC, and local biodiversity projects and centers will further promote the new data culture. We will create a feature story "Where do plant species occur?" for ASU's popular "Ask A Biologist" website, and a series of undergraduate student-led "How-To" videos that illustrate the use case workflows, including the creation of multi-taxonomy alignments.

Formal development of multi-agent systems using MAZE

MAZE is an extension of the Object-Z specification language supporting the specification and development of multi-agent systems (MAS). Following recommendations from the agent-oriented software engineering community, it supports three distinct levels of abstraction: (i) the macro level which focusses on the system's overall, global behaviour, independently of how the agents of the system operate and interact, (ii) the meso level which focusses on agent interactions, and (iii) the micro level which focusses on the operation of individual agents. Object-Z's high-level support for component-based specification, which is well suited to modelling MAS, is complemented in MAZE with support for action refinement to facilitate the top-down development process from the macro to micro level, and with a number of syntactic conventions aimed at abstractly specifying the low-level mechanisms required for dealing with asynchronous communication and timing constraints at the micro level. The latter are shorthands for existing Object-Z notation and so require no redefinition of Object-Z's semantics. In this paper, we provide an overview of MAZE and illustrate its use on a non-trivial case study: a swarm robotic algorithm for self-assembly.

VADER: A Parsimonious Rule-Based Model for Sentiment Analysis of Social Media Text

The inherent nature of social media content poses serious challenges to practical applications of sentiment analysis. We present VADER, a simple rule-based model for general sentiment analysis, and compare its effectiveness to eleven typical state-of-practice benchmarks including LIWC, ANEW, the General Inquirer, SentiWordNet, and machine learning oriented techniques relying on Naive Bayes, Maximum Entropy, and Support Vector Machine (SVM) algorithms. Using a combination of qualitative and quantitative methods, we first construct and empirically validate a gold-standard list of lexical features (along with their associated sentiment intensity measures) which are specifically attuned to sentiment in microblog-like contexts. We then combine these lexical features with consideration for five general rules that embody grammatical and syntactical conventions for expressing and emphasizing sentiment intensity. Interestingly, using our parsimonious rule-based model to assess the sentiment of tweets, we find that VADER outperforms individual human raters (F1 Classification Accuracy = 0.96 and 0.84, respectively), and generalizes more favorably across contexts than any of our benchmarks.

Analysis of short texts on the Web: introduction to special issue

1 Analysis of short texts on the Web Analysis of web and social media data is a rapidly growing area of research. Researchers seek to extract a wide variety of information from these texts in order to address specific user needs, profile attitudes and intentions, and target advertising, etc., which may require application of the full range of natural processing techniques. However, many of the texts in question—including news feeds, document titles, FAQs, and tweets—exist as short, sometimes barely sentence-like snippets that do not always follow the lexical and syntactic conventions assumed by many language processing tools. Many NLP analyses rely on the repetition of specific lexical items throughout the text in order to identify topic, genre, and other features; without sufficient context to enable such analyses, and because of their often eccentric grammatical style, short texts pose a new kind of challenge for

GLOBALIZATION, STANDARDIZATION, AND DIALECT LEVELING IN IRAN

This paper is an attempt to shed light on the effects of modernization, urbanization, monolingual educational system, and mass media as well as the process of globalization on dialect leveling among Persian dialects. In so doing, the first part of the paper elaborates on the relationship between globalization and sociolinguistics, and on the concept of standardization. Also, it discusses some factors accelerating dialect leveling among Persian dialects. The second part of the paper presents some empirical evidence for the change, based on the data collected from 3 Persian dialects, namely: Birjandi, Neyshaboori, and Yazdi. Three spoken language sample corpora from two age cohort groups in each speech community were collected. Having transcribed the data, the researchers juxtaposed the phonological, morphological, and syntactic conventions of the two groups of each speech community for further in-depth analysis. Comparing the speech samples of the two age cohorts in each corpus revealed significant movements towards standard dialect. However, the quality and quantity of the observed movements did not appear to be the same across the dialects in question.

Co-evolution of lexical and syntactic classifiers during a language game

This paper demonstrates for the first time that a learning classifier system can act as the core of a cognitive architecture to allow agents to co-evolve lexical and syntactic conventions for the efficient communication of conceptual strings during a language game. The use of the learning classifier system has several inherent advantages to previous ad-hoc approaches to modeling language games. It permits incremental addition of new populations of classifiers (e.g. syntactic classifiers) to modify communicative conventions as necessary without destroying or catastrophically interfering with lower level (e.g. lexical) conventions. The tendency for learning classifier systems to produce maximally general classifiers automatically produces systematic syntactic conventions. These results provide a proof in principle that replication of classifiers in the brain could serve an important role in cognition as proposed previously by the neuronal replicator hypothesis.

SYSTEMATIC CONSTRUCTION OF i* STRATEGIC DEPENDENCY MODELS FOR SOCIO-TECHNICAL SYSTEMS

Goal- and agent-oriented models have become a consolidated type of artifact in various software and knowledge engineering activities. Several languages exist for representing such type of models but there is a lack of associated methodologies for guiding their construction up to the necessary level of detail. In this paper we present RiSD, a method for building Strategic Dependency (SD) models in the i* notation. RiSD is defined in a prescriptive way to reduce uncertainness when constructing the model. RiSD tackles three fundamental issues: (1) it tends to reduce the average size of the resulting models; (2) it defines some traceability relationships among model elements; (3) it provides some lexical and syntactical conventions. As a result, we may say that RiSD supports the construction process of goal- and agent-oriented models whilst increasing their understanding.

Visual Aspects of Media Literacy

A central component of media literacy should be an understanding of the representational conventions through which the users of media create and share meanings. This paper analyzes the representational conventions of visual communications. It distinguishes between semantic and syntactic conventions and focuses on those characteristics that most sharply differentiate visual language from other modes of communication. It examines the impact of visual literacy on viewers' cognitive growth and on their development as informed consumers of visual media.

Multimedia interchange using SGML/HyTime. I. Structures

HyTime is a standard neutral markup language for representing hypertext, multimedia, hypermedia and time-based documents in terms of their logical structure. Documents represented in HyTime conform fully to the ISO Standard Generalized Markup Language (SGML). In effect, HyTime extends SGML by adding certain syntactic conventions called SGML architectural forms, with which it represents certain semantic constructs. HyTime cannot be understood or evaluated without understanding, at least to some extent, the significance and usefulness of the SGML standard on which it is based. This article examines the structure of the two standards. HyTime places unprecedented demands on document processing systems-demands which they have yet to meet. For example, a full implementation of HyTime would allow one to create a hyperlink to whatever happens to be going on at a particular time and/or place in a finite coordinate space (FCS), even if the event, location and time are not yet known, because of binding-time issues. HyTime allows a document to specify where and/or when the results of traversing a hyperlink will be rendered. HyTime provides constructs for specifying how events scheduled in one FCS are to be projected onto another, e.g. from a 3D FCS to a 2D FCS, or from a virtual measurement domain to a real one. >

Literalidad y literariedad en la traducción de la poesía de e.e. cummings

In this paper we attempt to prove that a reasonably literal translation of e.e. cummings is both more literary and adequate than a free one. Since cummings' poems are a paradigm of literariness achieved by means of the transgression of the norm, we believe that his break with typographical, syntactic and morphological conventions must be maintained in their translation. We discuss the danger of correcting cummings' transgressions in our translations for fear of the effects those "errors" might have on the reader. We also warn against giving an interpretation of those transgressions rather than a translation, as an interpretation will eliminate the effect of the radicality of these poems, will restrict the multiplicity of connotations of the original, and will rob the reader of the pleasure of making sense out of a poem that sounds hermetic at a first reading. Finally, we propose translations for two very different poems with the same subject matter: spring, and we discuss the reasons for our choice of options.

Formulaic Openings in Debussy

By now it comes as no surprise that much of what may at first strike us as innovative in Debussy's music was in fact carried out within the framework of historical precedent and established convention. Indeed, tracing the emergence of his personal style from the early works, with their explicit reliance on existing models, to the later works, where the models or conventions are more tacit, is an issue of fundamental musicological importance. Perhaps no composer of this fin-de-sikcle era exemplifies more clearly the expressive tension that results from the forging of a creative language that simultaneously works within a received tradition and strives for radical originality. The question of the composer's development is particularly complex, because his music grows from not merely one, but several syntactical conventions. On the one hand, Debussy perceived the language of Wagner as the most progressive available style, with its motivic organization and transformation, harmonic and formal freedom, aperiodic construction, and explicit seriousness of purpose.' On the other, he inherited by culture and education the set of languages of such French contemporaries and immediate predecessors as Gounod, SaintSains, Faur6, Duparc, Chabrier, Franck, and Massenet-languages not untouched by the example of Wagner (and Liszt), but ones more conservative in their melodic emphasis and periodicity, even while they admitted considerable harmonic experimentation, particularly in the area of modality. Debussy's early works blend in differing proportions the existing German and French late-nineteenth-century conventions: appropriate responses from a student and young composer searching for a distinctive Notes for this article begin on page 57.

A comparison of LISP specifications of function definition and argument handling

This paper presents the differences in several LISP definitions regarding the interpretation of various function types (EXPR, FEXPR, LEXPR, NEXPR). The definitions are compared according to their syntactic conventions for specifying the function type and method of argument handling--whether arguments are evaluated or unevaluated and whether their number is fixed or variable. A summary figure is given in the Appendix.

Epistemology and Ethnographic Reality: A Trobriand Island Case Study

According to the French film critic, Andre Bazin, the invention of cinema, as well as other recording instruments such as the phonograph, perpetuated the 19th-century that reality the events of the world around us-could be captured and reproduced by means of a mechanical device. In fact, the history of film technology can be viewed as a continuous effort to make this original an actuality. In the study of film esthetics (by critics such as Bazin and Christian Metz) can be seen, however, the equally continuous struggle to come to terms with the possible distortions created by the camera lens. The tradition of realism in cinema throughout the 20th century has emerged time and again, taking many different forms. But each reemergence has been associated with specific theoretical premises concerning the mode and meaning of filmic representations. Curiously, yet understandably, ethnographic films have not been subject to the same philosophical and methodological scrutiny with respect to the representation and/or the construction of reality. The very premises of the ethnographic method appear to obviate the problem of representing and/or constructing filmic reality. Too often, however, these assumptions reduce reality to something like getting the correct ethnographic facts and focusing the camera in the right place. In many cases when these two criteria are fulfilled, the narration in the film then imposes an interpretation which has the authority of truth, but which in fact may be the filmmaker's perception of truth. Such interpretations can be found in classic films, such as the survival theme in The Hunters (John Marshall 1956) and the theme in Dead Birds (Robert Gardner 1961). Both Gardner and Marshall, each in his own way, created a theme and used it as the central focus of his film, In The Hunters, the commentary and the editing present the Bushmen as victims of a barren, harsh physical environment (see Heider's discussion 1976:31-32). The narration continually tells the audience what the hunters are thinking as they track a giraffe (whose own thoug ts are also conveyed to the audience). The troubled thoughts of hunters and hunted reenforce the already overstated emphasis on hardship and the never-ending search for food. (The more recent ethnographic work of Richard Lee and others effectively destroys the survival theme.) Similarly, Gardner organized his film on the Dani of West Irian around a Dani myth which in the film functions as the significant principle in the Dani cosmology. According to Heider (1972; 1976:33), however (who was part of the original film expedition), the Dani do not perceive death in the philosophical way presented in the film. The major themes in both films are the result of the filmmakers' perceptions of reality, but they exist in the films as ethnographic truth. Through a commentary that speaks for the subjects, or in the process of editing, or simply by focusing the camera in a particular way, a filmmaker's assumptions of reality become incorporated into the finished presentation. Obviously the problem of reality is not new to ethnographic films. But too often questions of reality are debated around the issue of the reliability or unreliability of ethnographic facts (see, e.g., Heider 1976:11-15). The problem, I suggest, is much broader and demands the recognition that theories of filmic reality are not only the concern of cinema filmmakers and critics. I argue that ethnographic filmmakers must consider seriously their own epistemologies regarding filmic reality. Epistemologies, either explicitly or implicitly defined, exist in ethnographic films just as particular epistemologies underlie the structure of written ethnographic texts. In written texts, however, we have syntactical conventions that cue the reader into observing the author's formulations as apart from other realities. Further, we write within a historical tradition of epistemologically constructed models. In the conventions of filmmak-

RAPT: A language for describing assemblies

In the paper we describe a language for instructing assembly robots. This language has been implemented in an experi‐mental form on the DEC system 10 computer at Edinburgh. Since assembly robots resemble numerically controlled machine tools, we chose to use the APT language as the syntactic basis for the new language so that people accustomed to numerical control could readily adapt to it. We shall refer to this language as RAPT ‐ (Robot APT). It should be noted that the syntactic and semantic components of RAPT are sufficiently independent for other syntactic conventions to be used with the RAPT semantic interpreter.

Semantic taxonomy

Interest in semantic taxonomy is being rekindled after having been for sometime under the cloud of both transformational and pre-transformational formalism. However, whether practiced now by European Structuralists (like Göran Hammarström), Interpretive Semanticists (like Jerrold Katz), or Generative Semanticists (like George Lakoff), this interest is neither very semantic nor very taxonomic. There is a fundamental difference between atomistic analysis and genus-species synthesis in every science - most obviously in biology, the most conspicuous and successful area of taxonomic work. When this fact is recognized, then the way is cleared for a significantly different approach to taxonomy in linguistics. In contrast to the entrenched atomism of almost all modern linguistics is the present attempt to construct a semantic taxonomy based upon the assertion as the minimum independent individual instance of discourse. There are six semantic kinds, or species, of assertion: Identity, Subclass-class, Individual-class, Agent-function, Object-function, Characteristic. These combine into genus complexes by means of definite conjunctional and syntactical conventions. The four semantic kinds of inter-assertional relationships are: Continue, Contrast, Conclude, Support.