External Information Sources Research Articles

Aspects of Socio-Economic Success in a Labrador Village

During the last generation, the people of southern Labrador have enjoyed increasing contact with a number of sources of external information and, consequently, have developed middle class Canadian and American consumer desires. The degree to which these desires can be satisfied only partially determines the status of families in the community studied. This paper concerns how the community limits the definition of “success”. First, I briefly sketch the history and development of the community since its settlement to show the nature of traditional and modern values. Then the current community concept of “success” is examined with respect to its material and social constraints.

Policy signaling and stock price synchronicity: Evidence from China

We investigate whether stock price synchronicity is associated with province-level policy support signaling in the case of China. China provides a unique opportunity to investigate the impacts of varied policy signaling across provinces within one national environment. We show that stock price synchronicity is higher for firms in provinces with explicit policy support signaling. Furthermore, more intensive policy support signaling results in increasing stock price synchronicity. Additional robustness tests and mechanism analyses also confirm our empirical results. Our conclusions convincingly show that province-level policy signaling is an important external information source that is valued by China’s investors.

Modeling and Recommending Open Source Licenses with findOSSLicense

Open source software is widely used in the software industry and the academia. Licenses applied to open source software provide the terms for its further use and distribution. Decisions regarding licensing for new software systems are essential for the system's future use. In this paper, we introduce findOSSLicense, a license recommender that guides users into choosing the appropriate open source license for their software under creation. We also introduce our license modeling concept that is used in the recommendation process. The license modeling captures the properties usually found in existing open source licenses following an analysis performed on license texts. The recommendation process of findOSSLicense is based on a hybrid recommender that uses constraint-based, content-based and collaborative filtering giving also space for flexibility in the use of the system by its end-users who can adapt some system properties. User input, but also external sources of information including existing open source projects, are considered for the creation of the recommendations, whereas licenses used in third party software employed in the software are examined on a limited basis. findOSSLicense has been evaluated with the participation of users of various expertise.

Augmented EHR: Enrichment of EHR with Contents from Semantic Web Sources

This work presents methods to combine data from the Semantic Web into existing EHRs, leading to an augmented EHR. An existing EHR extract is augmented by combining it with additional information from external sources, typically linked data sources. The starting point is a standardized EHR extract described by an archetype. The method consists of combining specific data from the original EHR with contents from the external information source by building a semantic representation, which is used to query the external source. The results are converted into a standardized EHR extract according to an archetype. This work sets the foundations to transform Semantic Web contents into normalized EHR extracts. Finally, to exemplify the approach, the work includes a practical use case in which the summarized EHR is augmented with drug–drug interactions and disease-related treatment information.

InferNER: an attentive model leveraging the sentence-level information for Named Entity Recognition in Microblogs

We investigate the problem of named entity recognition in the user-generated text such as social media posts. This task is rendered particularly difficult by the restricted length and limited grammatical coherence of this data type. Current state-of-the-art approaches rely on external sources such as gazetteers to alleviate some of these restrictions. We present a neural model able to outperform state of the art on this task without recurring to gazetteers or similar external sources of information. Our approach relies on word-, character-, and sentence-level information for NER in short-text. Social media posts like tweets often have associated images that may provide auxiliary context relevant to understand these texts. Hence, we also incorporate visual information and introduce an attention component which computes attention weight probabilities over textual and text-relevant visual contexts separately. Our model outperforms the current state of the art on various NER datasets. On WNUT 2016 and 2017, our model achieved 53.48\% and 50.52\% F1 score, respectively. With Multimodal model, our system also outperforms the current SOTA with an F1 score of 74\% on the multimodal dataset. Our evaluation further suggests that our model also goes beyond the current state-of-the-art on newswire data, hence corroborating its suitability for various NER tasks.

Friendly Fire

IoT platforms enable users to connect various smart devices and online services via reactive apps running on the cloud. These apps, often developed by third-parties, perform simple computations on data triggered by external information sources and actuate the results of computations on external information sinks. Recent research shows that unintended or malicious interactions between the different (even benign) apps of a user can cause severe security and safety risks. These works leverage program analysis techniques to build tools for unveiling unexpected interference across apps for specific use cases. Despite these initial efforts, we are still lacking a semantic framework for understanding interactions between IoT apps. The question of what security policy cross-app interference embodies remains largely unexplored. This article proposes a semantic framework capturing the essence of cross-app interactions in IoT platforms. The framework generalizes and connects syntactic enforcement mechanisms to bisimulation-based notions of security, thus providing a baseline for formulating soundness criteria of these enforcement mechanisms. Specifically, we present a calculus that models the behavioral semantics of a system of apps executing concurrently, and use it to define desirable semantic policies targeting the security and safety of IoT apps. To demonstrate the usefulness of our framework, we define and implement static analyses for enforcing cross-app security and safety, and prove them sound with respect to our semantic conditions. We also leverage real-world apps to validate the practical benefits of our tools based on the proposed enforcement mechanisms.

Los vínculos sociales y la adopción de buenas prácticas de producción entre productores de café en Zongolica, Veracruz, México

Se analizan vínculos sociales y su relación con la adopción de buenas prácticas agrícolas en café desde el enfoque de las redes sociales. Se aplicaron entrevistas semiestructuradas sobre aspectos socioeconómicos, adopción de buenas prácticas de producción y relaciones sociales a 67 cafeticultores. Los productores de café tienen superficies y rendimientos promedio de 2.4 ha y 1.9 t/ha-1 respectivamente. Las prácticas adoptadas están relacionadas con el manejo del cultivo, control de plagas y uso de nuevas variedades. Existe una baja densidad de vínculos que indican escasa conectividad entre productores, nodos de investigación y proveedores. La estructura de relaciones bajo el escenario actual dificulta el proceso de adopción de buenas prácticas agrícolas por la baja vinculación a fuentes externas de información.

Evaluation of Disaster Preparedness and Preparedness Behaviors among Pharmacists: A Cross-Sectional Study in Australia.

In a disaster aftermath, pharmacists have the potential to provide essential health services and contribute to the maintenance of the health and well-being of their community. Despite their importance in the health care system, little is known about the factors that affect pharmacists' disaster preparedness and associated behaviors. The goal of this study was to determine the factors that influence disaster preparedness behaviors and disaster preparedness of Australian pharmacists. A 70-question survey was developed from previous research findings. This survey was released online and registered Australian pharmacists were invited to participate. Multiple linear regression was used to determine the factors that influenced preparedness and preparedness behaviors among pharmacists. The final model of disaster preparedness indicated that 86.0% of variation in preparedness was explained by disaster experience, perceived knowledge and skills, colleague preparedness, perceived self-efficacy, previous preparedness behaviors, perceived potential disaster severity, and trust of external information sources. The final model of preparedness behaviors indicated that 71.1% of variation in previous preparedness behaviors can be explained by disaster experience, perceived institution responsibility, colleague preparedness, perceived likelihood of disaster, perceived professional responsibility, and years of practice as a pharmacist. This research is the first to explore the significant factors affecting preparedness behaviors and preparedness of Australian pharmacists for disasters. It begins to provide insight into potential critical gaps in current disaster preparedness behaviors and preparedness among pharmacists.

Machine Learning-Based Processing Proof-of-Concept Pipeline for Semi-Automatic Sentinel-2 Imagery Download, Cloudiness Filtering, Classifications, and Updates of Open Land Use/Land Cover Datasets

Land use and land cover are continuously changing in today’s world. Both domains, therefore, have to rely on updates of external information sources from which the relevant land use/land cover (classification) is extracted. Satellite images are frequent candidates due to their temporal and spatial resolution. On the contrary, the extraction of relevant land use/land cover information is demanding in terms of knowledge base and time. The presented approach offers a proof-of-concept machine-learning pipeline that takes care of the entire complex process in the following manner. The relevant Sentinel-2 images are obtained through the pipeline. Later, cloud masking is performed, including the linear interpolation of merged-feature time frames. Subsequently, four-dimensional arrays are created with all potential training data to become a basis for estimators from the scikit-learn library; the LightGBM estimator is then used. Finally, the classified content is applied to the open land use and open land cover databases. The verification of the provided experiment was conducted against detailed cadastral data, to which Shannon’s entropy was applied since the number of cadaster information classes was naturally consistent. The experiment showed a good overall accuracy (OA) of 85.9%. It yielded a classified land use/land cover map of the study area consisting of 7188 km2 in the southern part of the South Moravian Region in the Czech Republic. The developed proof-of-concept machine-learning pipeline is replicable to any other area of interest so far as the requirements for input data are met.

An Evaluation Methodology for Interactive Reinforcement Learning with Simulated Users.

Interactive reinforcement learning methods utilise an external information source to evaluate decisions and accelerate learning. Previous work has shown that human advice could significantly improve learning agents’ performance. When evaluating reinforcement learning algorithms, it is common to repeat experiments as parameters are altered or to gain a sufficient sample size. In this regard, to require human interaction every time an experiment is restarted is undesirable, particularly when the expense in doing so can be considerable. Additionally, reusing the same people for the experiment introduces bias, as they will learn the behaviour of the agent and the dynamics of the environment. This paper presents a methodology for evaluating interactive reinforcement learning agents by employing simulated users. Simulated users allow human knowledge, bias, and interaction to be simulated. The use of simulated users allows the development and testing of reinforcement learning agents, and can provide indicative results of agent performance under defined human constraints. While simulated users are no replacement for actual humans, they do offer an affordable and fast alternative for evaluative assisted agents. We introduce a method for performing a preliminary evaluation utilising simulated users to show how performance changes depending on the type of user assisting the agent. Moreover, we describe how human interaction may be simulated, and present an experiment illustrating the applicability of simulating users in evaluating agent performance when assisted by different types of trainers. Experimental results show that the use of this methodology allows for greater insight into the performance of interactive reinforcement learning agents when advised by different users. The use of simulated users with varying characteristics allows for evaluation of the impact of those characteristics on the behaviour of the learning agent.

An innovative pharmacology curriculum for medical students: promoting higher order cognition, learner-centered coaching, and constructive feedback through a social pedagogy framework

BackgroundOngoing developments in medical education recognize the move to curricula that support self-regulated learning processes, skills of thinking, and the ability to adapt and navigate uncertain situations as much as the knowledge base of learners. Difficulties encountered in pursuing this reform, especially for pharmacology, include the tendency of beginner learners not to ask higher-order questions and the potential incongruency between creating authentic spaces for self-directed learning and providing external expert guidance. We tested the feasibility of developing, implementing, and sustaining an innovative model of social pedagogy as a strategy to address these challenges.MethodsConstructivism, communities of practice, and networked learning theory were selected as lenses for development of the model. Three hundred sixty-five first-year medical students participated between 2014 and 2018; they were introduced to pharmacodynamics and pharmacokinetics via 15 online modules that each included: learning objectives, a clinical vignette, teaching video, cumulative concept map, and small group wiki assignment. Five-person communities organized around the 15 wiki assignments were a key component where learners answered asynchronous, case-based questions that touched iteratively on Bloom’s cognitive taxonomy levels. The social pedagogy model’s wiki assignments were explored using abductive qualitative data analysis.ResultsQualitative analysis revealed that learners acquired and applied a conceptual framework for approaching pharmacology as a discipline, and demonstrated adaptive mastery by evaluating and interacting competently with unfamiliar drug information. Learners and faculty acquired habits of self-directed assessment seeking and learner-centered coaching, respectively; specifically, the model taught learners to look outward to peers, faculty, and external sources of information for credible and constructive feedback, and that this feedback could be trusted as a basis to direct performance improvement. 82–94% of learners rated the social pedagogy-based curriculum valuable.ConclusionsThis social pedagogy model is agnostic with regard to pharmacology and type of health professional learner; therefore, we anticipate its benefits to be transferable to other disciplines.

ОБҐРУНТУВАННЯ РЕКОМЕНДАЦІЇ ЩОДО УДОСКОНАЛЕННЯ ТАКТИКИ БОЙОВОГО ЗАСТОСУВАННЯ РАКЕТНИХ КАТЕРІВ З УРАХУВАННЯМ ІСТОРИЧНОГО ДОСВІДУ

The article shows the urgency of using the missiles by corvettes of the Ukrainian Navy for naval purposes in modern conditions. The fighting and local military conflicts at sea of the last century are analyzed, where showed that ships with missile weapons are majorly important in order to achieve a victory over the enemy. Today, Ukraine and its Armed Forces are at the stage of undeclared war with the Russian Federation. Currently, the combat zone is Donbass, and the zones of increased tension are the Azov and Black Seas. Analysis of the peculiarities of a possible theater of operations, showing the relatively low defense capability of Ukraine to counter the threat from the sea and the important economic importance of the Black and Azov Seas for Ukraine, prove that in this direction, the aggressor can achieve the greatest strategic success. Analysis of recent events in the Black Sea, namely: the incident in the Kerch Bay, shelling of Ukrainian warships and aircrafts, constant provocative actions of the aggressor prove that the Black Sea is likely to become a new theater of operations. Combat actions on the high seas are virtually ruled out, as they have no tactical prospects. The main actions of our forces will be aimed at defending the territorial sea and the lines of our own coast. The Navy’s defense zone is likely to be a coastal strip up to 20 nautical miles wide, as well as the area around Zmiyny Island. In the north-western part of the Black Sea, coast of Ukraine is characterized by the presence of a large number of shallow estuaries, bays, etc., closed by queues, sandy shoals, areas unsuitable for navigation (Odesa, Dniester banks, Dnieper and Dniester estuaries, Tendriv, Kizhanka bays, the mouth of the Dunay and others). Such conditions favor the actions of small corvettes using «ambush» tactics. Therefore, the main tactical way of Ukrainian Navy corvettes action in the event of an armed conflict at sea may be the use of naval missile armament from «ambush», under aircrafts cover of Ukraine Air Forces with the usage of external sources of information on enemy location. Keywords: tactics, missile boat, naval target, missile weapon, sea battle, missile strike, combat application, combat effectiveness, missile weapon, anti-ship cruise missile.

On the Solvability and Number of Solutions of Production-Logical Equations in a Fuzzy LP-Structure

For the construction and study of formal models of intelligent information systems, algebraic methods are useful. One of the topical directions here is the production-type logical systems, which are widespread in computer science. In recent years, the author and his followers have been developing the algebraic theory of LP-structures (lattice production structures). It is designed to formalize and solve a number of knowledge management problems in production systems. The method of relevant backward inference (LP-inference) was also introduced and investigated, which significantly reduces the number of calls to external information sources in comparison with classical inference. Subsequently, the theory was generalized to speed up inference in distributed production systems. At the same time, modern intelligent systems are characterized by fuzzy knowledge and fuzzy reasoning. There­fore, there is a need to extend the theory of LP-structures to fuzzy production systems. This research was initiated in previous articles by the author. Some concepts are introduced that impart fuzziness to LP-structures, and certain properties of fuzzy LP-inference are established. In recent works, research results are presented that systematically generalize the theory of LP-structures to the case of fuzzy knowledge bases. The terminology of FLP-structures with fuzzy logical relation (Fuzzy LP-structures) is introduced, the main standard properties are proved. The present work complements the FLP-structure model by investigating a class of production-logical equations. Relevant inference ideas are based on it, reducing the number of calls to external sources of information. Methods for solving these equations are formulated. For the first time, questions about the existence and number of solutions have been resolved. Finding a solution to a production-logical equation corresponds to the backward fuzzy inference in a production system. The proved theorems can be used for software implementation of fuzzy LP-structures and corresponding optimization of fuzzy inference. Some ideas for this implementation are discussed.

Adaptive selection of classifiers for bug prediction: A large-scale empirical analysis of its performances and a benchmark study

Bug prediction aims at locating defective source code components relying on machine learning models. Although some previous work showed that selecting the machine-learning classifier is crucial, the results are contrasting. Therefore, several ensemble techniques, i.e., approaches able to mix the output of different classifiers, have been proposed. In this paper, we present a benchmark study in which we compare the performance of seven ensemble techniques on 21 open-source software projects. Our aim is twofold. On the one hand, we aim at bridging the limitations of previous empirical studies that compared the accuracy of ensemble approaches in bug prediction. On the other hand, our goal is to verify how ensemble techniques perform in different settings such as cross- and local-project defect prediction. Our empirical experimentation results show that ensemble techniques are not a silver bullet for bug prediction. In within-project bug prediction, using ensemble techniques improves the prediction performance with respect to the best stand-alone classifier. We confirm that the models based on Validation and Voting achieve slightly better results. However, they are similar to those obtained by other ensemble techniques. Identifying buggy classes using external sources of information is still an open problem. In this setting, the use of ensemble techniques does not provide evident benefits with respect to stand-alone classifiers. The statistical analysis highlights that local and global models are mostly equivalent in terms of performance. Only one ensemble technique ( i.e., ASCI) slightly exploits local learning to improve performance.

СОВРЕМЕННАЯ УЧЕБНАЯ ФРАЗЕОГРАФИЯ: ИННОВАЦИОННЫЕ ПАРАМЕТРЫ БУМАЖНОГО СЛОВАРЯ

The article aims to solve the problem of inducing interactivity in the text of a printed educational phraseological dictionary. The relevance of the problem is justified by the user request of today’s school students and confirmed by psychologists’ conclusions about the features of children’s memory and perception in the postindustrial infor-mation society. The features of innovative interactive dictionaries prepared by the Pskov Laboratory of Educational Lexicography are shown against the background of the achievements of Russian and foreign educational phraseography. The authors pre-sent to students the linguistic properties of idioms and their cultural background identi-fied by the methods of historical-etymological, functional-semantic and linguocultural analysis in a series of interactive phraseological dictionaries, including: an entertaining etymological dictionary, a dictionary-workshop, a plot phraseological dictionary, and a game dictionary-quest. This variety of phraseographic genres is explained by the need to form a genre perception of the text and the role of phraseology in this text, which is also facilitated by the implementation of the principle of interactivity in the construc-tion of dictionary entries. The text interactivity in educational phraseological dictionaries of all these genres is achieved through techniques of dialogization of a lexicographic text, its hypertext structuring and game manipulation. Methods of dialogization of a phraseographic text (lexical-syntactic and compositional) ensure the inclusion of the child-recipient in the study of the origin of phraseology together with the characters of plot dictionary entries and their authors, motivate the student to perform cognitive creative tasks and give real feedback with the authors of the dictionary in the format of e-mail communication. Hypertextuality as an integral property of an interactive dictionary, which also stimulates the reader’s cognitive activity, is implemented in the paper format of the dictionary by referring to various external sources of infor-mation, cross-references in dictionary entries, and the use of surprise navigators. Game technologies of a lexicographic text construction are used at all levels of phra-seological units’ representation – from including phraseological game transforma-tions in the characters’ dialogues, which is typical of modern speech, to structuring the entire information and cognitive field of the dictionary as a search and etymolo-gical game-quest. Thus, the feasibility of developing innovative educational dictiona-ries in accordance with the concept of their interactivity is proved: this will ensure the availability of the material and, at the same time, the completeness and academic reliability of the representation of phraseological units as units of language and facts of Russian culture.

Innovation among SMEs in Finland: The impact of stakeholder engagement and firm-level characteristics

PURPOSE: The aim of the paper is to analyse the importance of both internal capabilities (resources) and external information sources in implementing product, process, marketing and organizational innovations aiming to maximize firm competitive advantage and create value for stakeholders. Furthermore, in particular, we examine the role of public organizations, business networks, firm size, and the industry sector, in the emergence of different types of product, process, marketing, and organizational innovations. The research was based on the typology of innovation (product, process, marketing, and organizational) adopted by the OECD. METHODOLOGY: The paper is based on data from 389 SMEs located in Finland and describes the development of a model for testing the factors that increase the innovativeness of SMEs. The logistic regression model is used as a methodology. Findings: The results show that the creation of novel products, processes and marketing innovation is connected to various external sources of information, such as fairs, the media and the internet. Moreover, the relationship between internal capabilities such as the firm’s know-how increases the marketing and organizational innovativeness of SMEs. Our results demonstrated that the creation of product innovation is positively connected to manufacturing. Furthermore, we find that the creation of novel processes and organizational innovation is related to firm size, such that firms with fewer than 20 employees (smallest firms) were concentrated among non-innovators and companies with more than 20 employees were concentrated among innovators. IMPLICATIONS FOR THEORY AND PRACTICE: The contribution of our study is to analyse to what extent various types of innovation rely on specific information sources. This study also provides suggestions for practice and policymakers. Contrary to expectations regarding our findings, public support organizations were not statistically significant in any innovation model. Therefore, public support organizations should develop better mechanisms to find SMEs with strong motivations to develop new products and market opportunities. ORIGINALITY AND VALUE: This paper provides a new and topical viewpoint for the literature by examining the possible factors explaining the increase in SMEs’ likelihood of implementing product, process, marketing, and organizational innovations. Our study provides comprehensive information on how different stakeholders contribute to the emergence of SME innovation.

Methods for Solving Production-Logical Equations in a Fuzzy LP-Structure

Algebraic methods provide an effective formalism for constructing and researching models of a wide range of information systems, especially intelligent ones. This provision fully applies to the production-type logical systems widespread in computer science. In the last decade, the author has created an algebraic theory of LP-structures (lattice production structures), which makes it possible to effectively solve a number of important problems related to production systems. Such tasks include equivalent transformations, verification, minimization of knowledge bases, and acceleration of logical inference. In particular, the method of relevant backward inference (LP-inference) was introduced and investigated, which significantly reduces the number of calls to external sources of information. Subsequently, this theory was expanded to model of distributed production systems. An important property of modern intelligent systems is the fuzzy nature of knowledge and reasoning. Therefore, an urgent problem arises of extending the advantages of the theory of LP-structures to fuzzy production systems. The beginning of this direction was laid in the previous articles of the author. Concepts describing the fuzzy LP-structure were introduced, and some useful properties of fuzzy LP-inference were studied. In recent works, studies are pre­sented that systematically generalize the theory of LP-structures for managing fuzzy knowledge bases. The basic terminology of FLP-structures with a fuzzy logical relation (Fuzzy LP-structures) is introduced, the basic properties are proved — closedness, the existence of a canonical form and logical reduction. This work complements this model by defining and investigating the apparatus of production-logical equations in the FLP-structure. Methods for solving these equations are proposed and substantiated. Finding a solution to the production-logical equation corresponds to the backward fuzzy inference. The presented theorems provide a theo­retical basis for further advances in the field of optimization of fuzzy inference. As a continuation of the work, it is planned to consider questions about the direct solvability of simplified equations and the number of their solutions.

The role of external information sources in organizational innovation in Nigeria

Traditionally, the focus of strategic management studies has been on technological innovation. However, recent studies have shown rising importance of non-technological innovation to firms’ productivity, especially in developing country contexts. One limitation in empirical firm innovation analyses this study attempts to resolve, is the role of information sources in driving non-technological (organizational) innovation in a developing country context. Developing countries rely on knowledge from external sources to make up for their deficiencies in R&D capability to drive their innovation efforts. Using pooled, cross-sectional data from two rounds of Nigeria’s Innovation Surveys, this study assesses the role of external information sources in organizational innovation practices in Nigeria. A comparative analysis of the manufacturing and service sectors was also undertaken. Using a multivariate probit analysis, our results provide strong evidence of heterogeneity in firms’ information sources used in implementing organizational innovation practices in Nigeria. Most information sources which drive organizational innovation practices in manufacturing firms do not have an impact on service firms. The study recommends sector-specific polices to drive organizational innovation in developing countries.

Effect of R&D intensity on the innovative performance of manufacturing companies. Evidence from Ecuador, Peru and Chile

PurposeThis paper aims to examine the relationship between research and development (R&D) intensity and innovative performance and R&D intensity as a moderating variable in the relationship between sources of information and innovative performance.Design/methodology/approachThis is a quantitative, nonexperimental, cross-sectional study of the data collected from national surveys of innovation activities from Ecuador, Peru and Chile where the investigation was carried out. A bivariate probit regression was applied.FindingsThe results of the investigation pinpoint that R&D intensity is positively related to the innovation of products and processes in Ecuador and Peru. However, no relationship was found in Chile. As a moderating variable of the information sources (customers, suppliers and competitors), and the innovation of products and processes, it shows different results in the three countries examined.Research limitations/implicationsThis study contributes to the literature with evidence in countries with low rates of investment in R&D in the countries examined, this relationship does not always exist; this relationship is considered to be dependent on the complexity of the knowledge and internal capabilities of the company required to achieve innovation, and this complexity could vary according to the type of manufacturing and technology level of the companies. Thus, in manufacturing companies of less complexity to achieve the necessary knowledge for innovation, low rates of investment in R&D are sufficient for the relationship to exist.Practical implicationsBy increasing their R&D intensity, companies acquire technology and develop internal skills and capabilities that boost their innovative potential. Nevertheless, it is not enough to increase R&D intensity to take advantage of external sources of information, it is also necessary to boost the absorptive capacity to assimilate and take advantage of external knowledge.Originality/valueThis study contributes to the scarce evidence that exists, on the literature in developing countries, on the effect of R&D intensity on innovative performance and provides evidence of R&D intensity as a moderating variable of the relationship between sources of information and innovative performance.

Mobile phones in the spread of unreliable information on Twitter: evidence from the 2017 French presidential campaign

This article analyzes the spread of unreliable information on Twitter during the 2017 French presidential campaign, focusing on the use of mobile phones with regard to information-sharing behavior. The corpus is composed of 38,346,765 tweets, posted by 2,163,812 supporters of the five main French political parties, from November 25, 2016 to May 12, 2017. We examine more precisely a sub-corpus of tweets (13,044,619) containing links to external information sources, in order to evaluate the different types of information sources and their reliability. Our research shows that information-sharing behavior within Twitter in France is generally based on reliable information sources, produced by journalists and professional media. However, we highlight that smartphone users tended to share a greater amount of user-generated content, as well as articles from a wider range of alternative political information sources (blogs, activists’ websites); such sources were most likely to publish unreliable information. Thus it appears that users of mobile phones tend to share more unreliable news than those who use Twitter from a computer web browser. Further, we show that this “device effect” on the spread of unreliable information is primarily amplified among the practices of one political community—namely, the far-right party and its network of supporters—which is more likely to organize debate around a larger number of unreliable references. We are claiming here that the design-based interoperability of these unreliable political news and social media applications helps to understand why the French far-right community shared more unreliable information from the Twitter application.