Keyword Filtering Research Articles

Using Twitter to Study Injection Site Reactions in Inflammatory Bowel Disease Patients

Introduction: Injection site reactions (ISRs) are a known adverse event (AE) of injectable therapies in patients with Crohn's disease (CD) and ulcerative colitis (UC), but real-time and long-term study of this AE remains difficult. The Twitter Streaming Application Programming Interface (API) has been previously used to monitor AEs in other diseases. The Twitter API (TAPI) gives access to ˜1% of all tweets and provides for large-scale data collection. We used the TAPI to describe injection site reactions in IBD patients. Methods: The Twitter Spritzer Streaming API and Search API were utilized together in order to collect tweets relevant to IBD and subcutaneous medications (adalimumab [ADA], golimumab, certolizumab pegol [CZP], and methotrexate [MTX]). User IDs were extracted to determine unique users. Retweets from all users and tweets from pharmaceutical companies were excluded. Disease information was extracted from tweets.Figure 1Results: Over two months, 220 tweets were collected from 104 unique users. 138 tweets (63%) were sent to Twitter from pharmaceutical companies and excluded. 54/220 tweets from 38 unique users (37%) met inclusion criteria (a 25% yield). Of these users, 27 were CD patients, 5 were UC patients, and 6 did not specify their type of IBD. The majority of Twitter users reported the use of ADA (n=28), 2 reported using CZP, and 8 reported using MTX. All users reported medication use by brand name except when describing MTX. 20 tweets (37%) contained information about time points. This included referencing injection schedules (n= 15), medication shipment dates (n=1), and changes in therapy (n=4). Two ISRs were reported, both in the use of ADA. The first ISR occurred in an individual with UC during the initial loading dose and was characterized by pain in their thighs for the duration of the day. No photo was included. The second occurred in a user with unspecified IBD and described slight itching and a reaction duration of greater than two weeks. A photo was uploaded confirming the ISR (Figure). Conclusion: Using Twitter to study injection site reactions identified a low number of reported AEs of interest, but provides a framework for directed inquiry for future use. The Streaming API can target these individuals and record important time points, changes in therapy, and use brand names as keyword filters.

Window of Opportunity for New Disease Surveillance: Developing Keyword Lists for Monitoring Mental Health and Injury Through Syndromic Surveillance.

Studies have documented the impact that hurricanes have on mental health and injury rates before, during, and after the event. Since timely tracking of these disease patterns is crucial to disaster planning, response, and recovery, syndromic surveillance keyword filters were developed by the New York State Department of Health to study the short- and long-term impacts of Hurricane Sandy. Emergency department syndromic surveillance is recognized as a valuable tool for informing public health activities during and immediately following a disaster. Data typically consist of daily visit reports from hospital emergency departments (EDs) of basic patient data and free-text chief complaints. To develop keyword lists, comparisons were made with existing CDC categories and then integrated with lists from the New York City and New Jersey health departments in a collaborative effort. Two comprehensive lists were developed, each containing multiple subcategories and over 100 keywords for both mental health and injury. The data classifiers using these keywords were used to assess impacts of Sandy on mental health and injuries in New York State. The lists will be validated by comparing the ED chief complaint keyword with the final ICD diagnosis code. (Disaster Med Public Health Preparedness. 2017;11:173-178).

The focus and frontier of corporate social responsibility

Purpose This paper aims to help scholars know about the focus and frontier in the field of corporate social responsibility (CSR). Although related research in CSR started 60 years ago, there is not much systematical literature review on CSR in recent years. This paper applies scientometric method, especially co-word analysis, to explore the frontier and focus of CSR in the twenty-first century, based on the articles from 2001 to 2014 in SSCI database. Design/methodology/approach In this paper, the authors first use the scientometric method and co-word analysis for keywords filtering and apply social network methodology to investigate the networks of high-frequency keywords and high-frequency authors. Findings The authors summarize the current state of CSR research in two aspects. By co-word analysis of high-frequency keywords, the findings show that the focus and frontier are highly related to CSR. Based on the findings of social network analysis, this paper concludes four important research directions and possible future research of CSR. Originality/value The findings in this paper will help scholars of CSR or other related fields to realize the focus and frontier of CSR and provide some guidance for their future research.

Big Data, Big Noise

In this article, we focus on noise in the sense of irrelevant information in a data set as a specific methodological challenge of web research in the era of big data. We empirically evaluate several methods for filtering hyperlink networks in order to reconstruct networks that contain only webpages that deal with a particular issue. The test corpus of webpages was collected from hyperlink networks on the issue of food safety in the United States and Germany. We applied three filtering strategies and evaluated their performance to exclude irrelevant content from the networks: keyword filtering, automated document classification with a machine-learning algorithm, and extraction of core networks with network-analytical measures. Keyword filtering and automated classification of webpages were the most effective methods for reducing noise, whereas extracting a core network did not yield satisfying results for this case.

Keyword Filtering Algorithm and Its Application in Polar Expedition Information Portal

In order to identify the basic data dictionary of polar expedition information portal relating to web texts, the first character searching and the attribute matching algorithms are proposed through analyzing text structure and characteristics of the basic data dictionary, and a method to look for basic data is provided based on these algorithms, which divides the whole basic data dictionary into two parts, i.e., first character filtering and first character associated filtering, thus this paper establishes a two-layer filtering model. It is used to identify the first character matching in the text and determine the association degree between different phrases or sentences in the basic data dictionary through attribute matching algorithm. The experimental results show that compared with traditional direct search algorithm, this first character searching and attribute matching algorithm could not only look for gazetteer more finely from various levels, but also effectively improve the accuracy of looking for basic data.

Su1430 Acute Effects of Botulinum Toxin on Pyloric Pressure and Distensibility in Patients With Pylorospasm

Most controlled in report negative results when compared to appropriate controls, which stands in stark contrast to the generally positive results of open label studies. We hypothesized that these effects are at least in part due to effects of patient selection with regression to themean. METHODS: Articles published between 1990 and 2014were searched in pubmed using the keyword gastroparesis and filters for human clinical trials and English. Articles were included if they described original data with patient-based symptom rating (mean plus SD, SEM or confidence intervals) obtained at two different time points separated by a standardized treatment. Data are given as mean with 95% confidence interval. RESULTS: A total of 199 publications were identified with 24 meeting inclusion criteria. A composite score for dyspeptic symptoms was provided in 19 studies (4 botulinum toxin [BX], 8 gastric electrical stimulation [GES], 7 medical therapy [MED]; placebo [PLA]). When normalized to a scale of 10, baseline severity differed significantly with BX (6.59 [5.55 7.63]) and GES (6.26 [5.49-7.03]) having higher severity score compared to MED (4.74 [4.26-5.22]) or PLA (5.17 [4.82-5.51]; P<0.05). Severity scores decreased with all treatments. Again BX 1.93 [0.94-2.94] and GES 3.15 [2.45-3.85] showed greater changes, compared to MED 1.85 [1.3-2.41] and PLA (2.15 [1.76-2.55]); P<0.05. Meta-regression demonstrated

Going Global

Going Global

Bulk Construction of Geo-Textual Indices

With the proliferation of online web objects, spatial keyword queries that retrieve objects satisfying both spatial and keyword conditions are becoming popular. The geo-textual index that integrates a spatial index with a keyword filter offers a promising approach for processing such queries efficiently. However, it is still an open problem on how the geo-textual indices can be effectively constructed from scratch. This paper proposes a bulk loading algorithm for the geo-textual indices, which constructs the indices top-down efficiently. It utilizes a two-phase method to construct the children of nodes at each level of the geo-textual index, taking both spatial and keyword information into consideration, and thus optimizes the geo-textual index for spatial keyword queries. The time complexity of the algorithm is analyzed theoretically, and comprehensive experiments with both real and synthetic datasets show that it can build high quality geo-textual indices with good query performance and space utilization.

Slovene Ethnographic Museum Web Collections

EXTENDED ABSTRACT:Are you interested in the night view of Ljubljana or in the view on the Ljubljana skyscraper in the 30- ies of the 20th century? What did Cerknica look like in that time? What did a typical alpine house look like? What beds were used fifty years ago? Are you a sports fan interested in the training of Yugoslav athletic team in 1953? Is it true that Anton Codelli and his colleague Leo Poljanec set up the first wireless telegraph station for the company Telefunken in the West African German colony of Togo? What is a Palm Sunday bundle? Is a duel between a woman of Carinthia and Carniola really painted on a wooden bee hive panel? What on earth is an object called »roš«? What chimneys were built on the roofs of the Artiže village? The answers to these unusual questions can be found on the SEM (Slovene Ethnographic Museum) website. The SEM documentation comprises numerous collections of photographs which are of interest to ethnologists as well as to the general public. To facilitate the access to them SEM bought the first scanner and started to scan individual photos in 1997. In 2007 SEM continued with systematic digitisation of the most frequently used collections of photos. In 2009 the activities were continued (partly in cooperation with external partners) and we joined the Athena EU project. In order to be promoted the entire collections were published on the SEM website and made accessible to a broad audience. The MINOK software enabling the creation of html catalogues was used. As static catalogues were unsuitable for inclusion on the website, a new application using dynamic html was designed to create the gallery view of museum objects and photos, photo slide show, keyword filter, geolocation filter, display of geolocation on a map, and to offer the possibility of advanced searching. The Galerist software was designed by an external partner. It enables html catalogue import, sorting and editing XML files; it is possible to add geolocation and to create specific collections etc. The Galerist web application was integrated to the SEM website which is built on the Drupal open source content management system (CMS). As the two search engines were not linked together, users had to make two separate searches which proved to be inconvenient. Many website visitors used the Drupal search engine instead of the Galerist. As the Drupal search engine is not convenient for simple searching, we decided to import the entire collection to the Drupal to facilitate the searching and to enable users to search the Galerist and the Drupal simultaneously. SEM collections were published on the website in 2010. They are divided into Collections of objects (e.g. photos of Palm Sunday bundles, the Kos collection, Folk instruments, Ploughs, Painted beehive panels, Headwear, Plastic carnival masks, Tally sticks, Blue-printing models, Superstition, Ploughing implements, the Milan Kovač collection and Horsehair sieves. They can be accessed at http://www.etno-muzej.si/sl/zbirke-predmetov) and Collections of photos (photos of 32 terrains shot by the teams under the guidance of dr. Boris Orel and other individual albums: Vekoslav Kramarič, Rado Kregar, Matija Murko, Slavko Smolej, Veno Pilon, Jernej Šušteršič, Press Office, Togo Album, Fran Vesel, Collection of old photos and Collection of Marko Račič' sports photos. They can be accessed at: http://www.etno-muzej.si/sl/zbirke-fotografij). The web collection comprises 55 collections with more than 23,000 entries. On the basis of Web statistics and users’ feedback we can assume that our collections are of interest to the general public. The primary goal to promote our cultural heritage was attained as people are generally acquainted with our permanent exhibitions but less familiar with our collections. Digitisation was funded by the EU Athena project - our collections are accessible via the Europeana website. SEM web collections project has not been concluded yet and is being constantly upgraded.

Monitoring Influenza Epidemics in China with Search Query from Baidu

Several approaches have been proposed for near real-time detection and prediction of the spread of influenza. These include search query data for influenza-related terms, which has been explored as a tool for augmenting traditional surveillance methods. In this paper, we present a method that uses Internet search query data from Baidu to model and monitor influenza activity in China. The objectives of the study are to present a comprehensive technique for: (i) keyword selection, (ii) keyword filtering, (iii) index composition and (iv) modeling and detection of influenza activity in China. Sequential time-series for the selected composite keyword index is significantly correlated with Chinese influenza case data. In addition, one-month ahead prediction of influenza cases for the first eight months of 2012 has a mean absolute percent error less than 11%. To our knowledge, this is the first study on the use of search query data from Baidu in conjunction with this approach for estimation of influenza activity in China.

Design and Implementation of Keyword Filter System Based on Netfilter

A keyword filter system is constructed based on Linux kernel firewall and netfilter-queue. Packets with http information will be sent to Netfilter and wait for decisions through using IP tables. The system can do deep filtering to packets and use multi-pattern algorithm, which increases the fitness and efficiency.

Survey of Keyword Filtering for Anti-Active Jamming in Chinese

This paper made an induction from case of keyword filtering based on anti-active jamming in Chinese. The main work was summarized related works of current situation and technology level, and then analyzed existed problems for traditional information filtering, at last was the trend of forecasting for keyword filtering based on anti-active jamming in Chinese.

Review: Finishing and polishing procedures of (resin-modified) glass ionomers and compomers in paediatric dentistry

A smooth surface has a beneficial effect on the aesthetic quality and longevity of a dental restoration, as well as on its biocompatibility with oral tissues. In this review studies on glass-ionomer cements (GIC), resin-modified glass-ionomer cements (RM-GIC), and compomers or polyacid-modified resin composites (PAM-C) were assessed as these are currently used in paediatric dentistry. Medline databases (USA National Library of Medicine) was searched using WinSpirs and Pubmed. This search used a keyword filter including dental materials, polishing procedures, and instrumentation including influencing factors. Although the literature shows different commercially available tools it seems that the ideal polishing instrumentation in paediatric dentistry for GIC and compomers is not yet on the market. Furthermore, the constant development of dental materials and polishing products impose a need for continuous scientific research.

A Database Grid System for Multi-Domain Dynamic Data Integration

With the richness of common database resources, distributed users in wide areas hope to transparently access and use these data resources on demand. DS_Grid (database grid) is an SOA (service-oriented architecture) based database grid system for data sharing in multiple application domains. DS_Grid adopts a P2P (peer-to-peer) Multi-Chord (MultiChord) architecture to realize the distributed storage, query processing and dynamic data integration of data resources. According to the text similarity, the data resources are registered to the corresponding domains to realize rapidly discovering data resources. The domain ontology knowledge and reasoning rules are used to support the semantics based intelligent query. A multi-root and multi-peer maintenance based data resources replica management mechanism is applied to improve the reliability of the system. A keyword filter based distributed data integration strategy is adopted to reduce the communication cost. A distributed clustering technique is used to summarize the huge data information. The experiments demonstrate the feasibility and effectiveness of the key techniques of DS_Grid.

APPLICATION OF CONCEPTUAL LEARNING TECHNIQUES TO GENERALIZED GROUP TECHNOLOGY

Abstract In this article, we first point out the advantages and limitations of the keywords filtering techniques and the standard automatic clustering techniques for group technology applications. We consider, then, a product, or more generally an item, as a conjunction of symbolic attributes. We present a new method to recognize homogeneous families of products and to interpret the families by extraction of their concepts. Finally, the method is applied to analyze a population of 1010 mechanical parts.