Digital Mixed Content Analysis on Digital Platform Social Data

The explosion of platform social data as digital secondary data, collectible through sophisticated and automatized query systems or algorithms, makes it possible to accumulate huge amounts of dense and miscellaneous data. The challenge for social researchers becomes how to extract meaning and not only trends in a quantitative as well as in a qualitative manner. Through the application of a digital mixed content analysis perspective to data analysis, in this contribution, the author will present the potentiality of a hybrid digitalized approach to social content. This perspective should be seen as an applied example of organizing a framework to guide the application of integrated methods of content analysis (quantitative and qualitative) but also integrated objects of analysis (individuals, relationships, and digital actions) on digital platform social data and to address their varied nature.

Match Algorithms for Scientific Names in FlorItaly, the Portal to the Flora of Italy

Scientific names are not part of everyday language in any modern country, and their input as strings in a query system can be easily associated with typographical errors. While globally unique identifiers univocally address a taxon name, they can hardly be used for querying a database manually. Thus, matching algorithms are often used to overcome misspelled names in query systems in several data repositories worldwide. In order to improve users’ experience in the use of FlorItaly, the Portal to the Flora of Italy, a near match algorithm to resolve misspelled scientific names has been integrated in the query systems. In addition, a novel tool in FlorItaly, capable of rapidly aligning any list of names to the nomenclatural backbone provided by the national checklists, has been developed. This manuscript aims at describing the potential of these new tools.

From Specialized Web Corpora of Tourism to a Learner’s Dictionary

This paper presents the two approaches used in creating specialized web corpora of Croatian tourism in Japanese for their usage in building a specialized learners’ dictionary. Both approaches use the WebBootCat technology (Baroni et al. 2006, Kilgarriff et al. 2014) to automatically create specialized web corpora. The first approach creates the corpora from the selected seed words most relevant to the topic. The second approach specifies a number of web pages that cover tourism-oriented information on specified regions, cities, and sites in Croatia available in Japanese, which are then used for web corpora creation inside the Sketch Engine platform. Both approaches provide specialized web corpora small in size, but quite useful for lexical profiling in the specific field of tourism. In the process of dictionary creation, the second approach has proven to be especially useful for the selection of lexical items, while both approaches have proven to be highly useful for the exploration and selection of authentic examples from the corpora. The research exposes some shortcomings in Japanese language processing, such as errors in the lemmatization of some culturally specific terms and indicates the need to refine existing language processing tools in Japanese. The Japanese-Croatian bilingual learner’s dictionary (Srdanović 2018) is currently in the pilot phase and is being used and built by learners and teachers through the open-source dictionary platform Lexonomy (Mechura 2017). In addition to the fact that work on the bilingual dictionary is useful as a means for training students in language analysis and description using modern technologies (e.g. corpora, corpus query systems, dictionary editing platform), the dictionary is also important in educating new personnel capable of working in tourism using the Japanese language, which is strongly needed. In future, the same approach could be used for creating specialized corpora and dictionaries for Japanese and other language pairs.

Barriers State Agencies Face with the Dissemination of Local Health Data through Web-Based Data Query Systems (Preprint)

BACKGROUND Web Based Data Query Systems (WDQS) make health data at the local level easily accessible to the public health. Despite their benefits Many state and local health agencies face significant challenges with their dissemination. OBJECTIVE The purpose of this study is to identify the most significant challenges they face from the perspective of Behavioral Risk Factor Surveillance System (BRFSS) coordinators. We also seek to find an association between perceived system aspects, challenges faced, contextual factors, and overall satisfaction with state level health data systems. METHODS We surveyed Behavioral Risk Surveillance System (BRFSS) coordinators from 43 states. We surveyed participants about contextual factors and asked them to rate system aspects and challenges they face with their health data system on a Likert scale. We used two sample t-tests to compare means on participant ratings for states with and without Web Based Data Query Systems (WDQS). RESULTS Overall, 95.4% of states make health data available over the internet, while 65.1% employ a WDQS. States reported the challenge of cost of hardware/software as a greater challenge between states with WDQS than without WDQS. States rated standardization of vocabulary more favorably in states with WDQS (n=3.32; 95% CI, 2.94-3.69) versus states without WDQS (n=2.85, 95% CI, 2.47-3.22). CONCLUSIONS Securing adequate resources, and commitment to standardization are vital in the dissemination of local level health data. Factors such a receiving data in a timely manner, privacy, and political opposition are less significant of a barrier than anticipated.

Improving the Reliability of Web Based Result Query Systems during High Traffic Periods

Displaying of examination results by a single central entity, for lakhs of students becomes a tedious task, and sometimes may also result in server crashing. These servers typically rely on heavy and often unrestricted threads spawned to handle each incoming request which is the reason why the server resources are used up quickly. We propose a solution that is three fold: First, multiple Volunteer entities are brought in to hold the data and donate a portion of their computing power to offload the enormous work placed on the central entity. Second, the central entity is changed to play the role of dispatcher that generates monitors and assigns extremely lightweight, independent processes (called agents) to each user request without requiring any additional hardware upgrade. Each agent will be responsible to satisfy their assigned user requests. Third, we introduce a load balancing technique derived from the ideas of autonomous agents load balancing techniques in cloud to provide load balancing among the Volunteer entities and the central entity such that the Volunteer entities can continue with its own tasks and not be overwhelmed by its Volunteer work while ensuring fast response time and better reliability and response to the user.