Online Information Extraction

A few years back we embarked on an expedition into the rapidly transforming landscape of data research, the narratives of big data and the practices emerging with novel data resources, tools and new directions of social and cultural inquiry. This book represents our own experiences and impressions of this journey. We were out there on unfamiliar and at times even uncharted territory. Often we depended on the help of more learned colleagues who generously shared their knowledge, gave us directions or helped with problems. Many of them became authors for this book and we would like to thank them for trusting us with collecting their contributions and presenting them in a joint volume. The editors wish to thank the Utrecht Data School staff Thomas Boeschoten, Irene Westra, Iris Muis, Daniela van Geenen and Gerwin van Schie for their input and for providing an intellectually stimulating work environment. With the Utrecht Data School we have created a place where ambitious and enthusiastic students can meet to join us in this exploration. We are grateful for having the rare opportunity of conducting research together with students from whom we can learn so much and whose insatiable curiosity is an inspiration as well as a constant reminder of why we became teachers in the first place. Our gratitude extends also to the Institute of Cultural Inquiry and the open access fund at Utrecht University for enabling us to make this book open access. A special thanks to William Uricchio, Fernando van der Vlist and Eef Masson for their helpful comments and advice at various stages of the editing process. Finally, we are particularly grateful to Nicolás López Coombs for helping us with the editing and the formatting of the book, but most importantly for keeping an eye on our timeline. Mirko Tobias Schäfer & Karin van Es Anywhere but in their office, 2016 This book is an important contribution towards meeting the challenges of the platform-driven, data-fuelled world in which we have all come to live.

This paper proposes a flow-batch methodology for the determination of free glycerol in biodiesel that is notably eco-friendly, since non-chemical reagents are used. Deionized water (the solvent) was used alone for glycerol (sample) extractions from the biodiesel. The same water was used to generate watercavitation sonoluminescence signals, which were modulated by the quenching effect associated with the amount of extracted glycerol. The necessarily reproducible signal generation was achieved by using a simple and inexpensive piezoelectric device. A linear response was observed for glycerol within the 0.001-100 mg/L range, equivalent to 0.004-400 mg/kg free glycerol in biodiesel. The lowest measurable concentration of free glycerol was estimated at 1.0 mg/L. The selectivity of the proposed method was confirmed by comparing the shape and retention of both real and calibration samples to standard solution chromatograms, presenting no peaks other than glycerol. All samples (after extraction) are greatly diluted; this minimizes (toward non-detectability) potential interference effects. The methodology was successfully applied to biodiesel analysis at a high sampling rate, with neither reagent nor solvent (other than water), and with minimum waste generation. The results agreed with the reference method (ASTM D6584-07), at a 95% confidence level.

This paper show how location named entity (LNE) extraction and annotation, which makes part of our named entity recognition (NER) systems, is an important task in managing the great amount of data. In this paper, we try to explain our linguistic approach in our rule-based LNE recognition and classification system based on syntactico-semantic patterns. To reach good results, we have taken into account morpho-syntactic information provided by morpho-syntactic analysis based on DIINAR database, and syntactico-semantic classification of both location name trigger words (TW) and extensions. Formally, different trigger word sense implies different syntactic entity structures. We also show the semantic data that our LNE recognition and classification system can provide to both information extraction (IE) and information retrieval(IR).The XML database output of the LNE system constituted an important resource for IE and IR. Future project will improve this processing output in order to expl...

Digital research is often understood as data-driven. Yet the ways in which data are already informed by specific analytical assumptions and inscriptions of the media in which they originate, circulate, or are being used is often neglected. The thesis intervenes in such focus on digital research in two ways. First, it offers a disciplinary contribution to the areas of digital social research and software studies through the development of a device-driven perspective. Such a perspective focuses on digital media both as object and as process, and is attentive to how digital media become operative in the digital research process. Second, a contribution to digital methods is made by introducing the notion of research affordances, which provide ways to operationalize the device-driven perspective and to discuss the implications of digital media as sources of data for the research process. The contributions are developed through empirical case studies examining the modes of research afford...

The paper studies the diversity of ways to express entity aspects in users’ reviews. Besides explicit aspect terms, it is possible to distinguish implicit aspect terms and sentiment facts. These subtypes of aspect terms were annotated during SentiRuEval evaluation of Russian sentiment analysis systems organized in 2014–2015. The created annotation gives the possibility to analyze the contribution of non-explicit aspects to the overall sentiment of a review, their main patterns, and possible use.

Abstrak Meningkatnya interaksi pengguna internet dan media sosial tentu memiliki dampak terhadap peningkatan jumlah data atau konten yang dihasilkan oleh pengguna. Data atau konten yang dihasilkan sering disebut dengan User Generated Content (UGC). Data UGC dapat dimanfaatkan untuk tujuan tercapainya strategi bisnis terhadap produk yang disediakan. Hal ini dilakukan dengan menganalisis interaksi yang terjadi dengan sebuah metode yaitu Social Network Analysis (SNA), dengan membandingkan properti jaringannya. Dalam penelitian ini dilakukan analisis interaksi di dalam platform jejaring sosial yaitu Twitter pada produk smartphone Xiaomi Redmi 9, Xiaomi Redmi 9A, Xiaomi Redmi 9C, Xiaomi Redmi Note 9, dan Xiaomi Redmi Note 9 Pro. Dari perbandingan properti jaringannya dihasilkan bahwa terdapat tiga tipe smartphone yang memiliki pertimbangan untuk dilakukan perencanaan pengadaan stok ke depannya, yaitu Redmi 9A, Redmi 9C dan Redmi Note 9 Pro. Kata kunci : User Generated Content, Social Net...

This study seeks to contribute to recent debates concerning computational social science by experimenting with ‘co‐occurrence analysis’ on a Twitter dataset relating to the subject of the recently introduced General Data Protection Regulation (GDPR). In so doing, the study implements the method in the context of hashtags associated with the subject, in an attempt to test the method’s capacities and limitations to detect liveliness ‐ dynamism and engagement with a given topic ‐ on Twitter. The study highlights the significance of a qualitative assessment of tweets in combination with network mapping. It further suggests that the ability of hashtag analysis to detect liveliness of issues on Twitter is influenced by the type of actors involved, insofar as the ubiquitous use of multiple hashtags in the same tweet as observed in this study poses serious limitations to hashtag analysis. In terms of specific hashtag analyses, the research stresses the importance of a comprehensive explorat...

There is a plethora of publications emerging in the humanities, especially media studies, that use data points from social media platforms in order to investigate social interaction and cultural production. Data points taken from social media platforms are used for calculating metrics on the most diverse aspects of users and use. As Carolin Gerlitz has pointed out, research practices tend to treat these data points alike in spite of the fact that they take on different functions (Gerlitz & Rieder 2013) and even though they are used by different social groups (Bruns & Stieglitz 2013). Drawing on Espeland & Stevens (1998), Gerlitz calls this the commensuration of data points and formulates a data point critique. What does your criticism of making data points countable and comparable for analysis consist of exactly?

This paper inquires in the politics of real-time in online media. It suggests that real-time cannot be accounted for as a universal temporal frame in which events happen, but explores the making of real-time from a device perspective focusing on the temporalities of platforms. Based on an empirical study exploring the pace at which various online media produce new content, we trace the different rhythms, patterns or tempos created by the interplay of devices, users' web activities and issues. What emerges are distinct forms of 'realtimeness' which are not external from but specific to devices organized through socio-technical arrangements and practices of use. Realtimeness thus unflattens more general accounts of the real-time web and research, and draws attention to the agencies built into specific platform temporalities and the political economies of making real-time.

Aquilo que muitos conhecem popularmente como Internet, caracteriza-se, socioculturalmente como ciberespaço e possui territorialidade própria, bem como suas próprias práticas culturais, identificadas como cibercultura. Tendo em vista que, para trazer à luz percepções fundamentais acerca de determinados territórios faz-se necessário analisá-los sob diferentes óticas, contemplando dimensões tais como a política, a econômica, a social e a cultural, este artigo tem como objetivo a realização de um estudo exploratório sobre a territorialidade digital do ciberespaço, por meio de pesquisa bibliográfica e análise de conteúdo, de modo a examinar e destacar aspectos socioculturais a partir das alterações promovidas pela Web 2.0 e percepções sobre as frações do ciberespaço designadas genericamente de Deep Webe Dark Web.

This is the first e-special issue for the journal Sociology and its chosen focus is the article 'The coming crisis of empirical sociology' by . This article challenged sociologists with a variety of questions about the role, relevance and methodological opportunities for sociological research in the 21st century. On publication it stoked the already charged debates on a public sociology (Burawoy, 2004), the role of publicly funded research (ESRC, 2009) and relevance of sociological research in an age of burgeoning social media . This e-special provides a reprise of these debates and explores relevant papers in Sociology, as well as alerting readers to recurring themes and new directions on the topic of methods and social research.

There is a need to study whether consumer trust and e-commerce information quality are the answers to the question what drives customers' purchase decision and consequently their satisfaction. This gap in knowledge can be a significant concern, especially for Indonesian e-commerce consumers. Hence, we gather 400 respondents to participate in our research. By conducting descriptive quantitative research with the assistance of AMOS version 21 software, we demonstrate important antecedents of customers' purchase decision and its subsequent effect on consumer satisfaction. Our results show that among Indonesian e-commerce consumers, information quality is slightly more important factors compared to consumer trust.

Qualitative and mixed methods digital social research often relies on gathering and storing social media data through the use of APIs (Application Programming Interfaces). In past years this has been relatively simple, with academic developers and researchers using APIs to access data and produce visualisations and analysis of social networks and issues. In recent years, API access has become increasingly restricted and regulated by corporations at the helm of social media networks. Facebook (the corporation) has restricted academic research access to Facebook (the social media platform) along with Instagram (a Facebook-owned social media platform). Instead, they have allowed access to sources where monetisation can easily occur, in particular, marketers and advertisers. This leaves academic researchers of digital social life in a difficult situation where API related research has been curtailed. In this paper we describe some rationales and methodologies for using APIs in social research. We then introduce some of the major events in academic API use that have led to the prohibitive situation researchers now find themselves in. Finally, we discuss the methodological and ethical issues this produces for researchers and, suggest some possible steps forward for API related research.

We respond to the two comments on our article `The Coming Crisis of Empirical Sociology' from Rosemary Crompton (2008) and Richard Webber (2009) which have been published in Sociology , as well as issues arising from the wider debate generated by our article. We urge sociologists to recognize the gravity of the challenges posed by the proliferation of social data and to become more vociferous in contributing to political debates over method and data.

As there is an enormous amount of online research material available, finding pertinent information for specific purposes has become a tedious chore. So there is a requirement of the research paper recommendation system to facilitate research scholars in finding their interested and relevant research papers. There are many paper recommendation systems available, most of them are depending on paper assemblage, references, user profile, mind maps. This information is generally not easily available. The majority of the prevailing recommender system is based on collaborative filtering that rely on other user's proclivity. On the other hand, content-based methods use information regarding an item itself to make a recommendation. In this paper, we present a research paper recommendation method that is based on single paper. Our method uses content-based recommendation approach that employs information extraction and text categorization.. It performs the profile learning by using naive Bayesian text classifier and generates recommendation on the basis of an individual's preference.

As social media technologies such as Twitter, Instagram, and YouTube have become highly ubiquitous, social life itself has become reconfigured. Though early notions of an offline/online binary remain in some quarters of social research, there is a realization amongst most that this binary is reified. As such, the study of social interactions within social media is a fundamental sociological question. This chapter argues that social researchers need to engage with the study of social media in order to comprehensively understand modern social life. This chapter also provides insights into how we, as social researchers, can critically collect and discern social formations via social media. Twitter is specifically used in this chapter to provide an example of how the medium provides opportunities for mixed qualitative and quantitative social analysis. Ultimately, this chapter also argues that the understanding of large social questions is increasingly contingent on us deciphering and understanding social knowledge formed and maintained within social media platforms.

The purpose of this paper is twofold: First, to suggest that techniques for mapping public disagreements over claims to knowledge, or controversies, can act as assistive devices for researchers in geography to move from research topics to research questions. A second purpose is to offer a recipe or ‘how-to’ guide of specific approaches and free scholarly software that researchers without specialized coding skills can use to achieve their own controversy mapping goals. We use a case of creating a controversy map covering the issue of transboundary movements of electronic waste (e-waste) to illustrate how these approaches and software can be associated together so other researchers may put them to use for their own controversy mapping purposes.

Digital Methods can be defined as the repurposing of the inscriptions generated by
digital media for the study of collective phenomena. The strength of these methods
comes from their capacity to take advantage of the data and computational capacities of
online platforms; their weakness comes from the difficulty to separate the phenomena
that they investigate from the features of the media in which they manifest (‘the medium
is the message’, according to McLuhan’s 1964 dictum). In this article, we discuss various
methodological difficulties deriving from the lack of separation between medium and
message and propose eight practical precautions to deal with it.

— Landslides are the most threatening geo-hazard. It is a kind of genetic type of slope and has same characteristics with slope. Chaotic time series of landslide displacement and its influential factors could reflect the history of landslide displacement of dynamic system. This study aims to investigate suitable model and predict future displacement values. Displacement could be predicted by considering and reconstructing the relationship among the triggering factors and LRN and BPNN could reflect relationship among variables. Temporal back propagation algorithm has been employed in traditional BPNN. Comparative analysis of the models with other existing techniques such as RBF shows that, RBF produces accurate results only during training but prediction error is not less than that of LRN and BPNN. Comparisons of all three models (RBF, LRN and BPNN) have been performed using MSE (Minimum Square Error) and regression values. Keywords— LRN, BPNN, prediction of landslide displacement, temporal back propagation learning.

The field of sentiment analysis, in which sentiment is gathered, analyzed, and aggregated from text, has seen a lot of attention in the last few years. The corresponding growth of the field has resulted in the emergence of various subareas, each addressing a different level of analysis or research question. This survey focuses on aspect-level sentiment analysis, where the goal is to find and aggregate sentiment on entities mentioned within documents or aspects of them. An in-depth overview of the current state-of-the-art is given, showing the tremendous progress that has already been made in finding both the target, which can be an entity as such, or some aspect of it, and the corresponding sentiment. Aspect-level sentiment analysis yields very finegrained sentiment information which can be useful for applications in various domains. Current solutions are categorized based on whether they provide a method for aspect detection, sentiment analysis, or both. Furthermore, a breakdown based on the type of algorithm used is provided. For each discussed study, the reported performance is included. To facilitate the quantitative evaluation of the various proposed methods, a call is made for the standardization of the evaluation methodology that includes the use of shared data sets. Semantically-rich concept-centric aspect-level sentiment analysis is discussed and identified as one of the most promising future research direction.

This paper describes the Information extraction and content
analysis system. The proposed system based on a conditional random eld algorithm and intended to extract aspect terms mentioned in the text. We used a set of morphological features for machine learning. The system was used to automatic extraction of explicit aspects and also to automatic extraction of all aspects (explicit, implicit and sentiment facts), and tested on two domains { restaurants and automobiles. We show that our system can produce quite a high level of precision which means that the system is capable of recognize of aspect terms rather accurate. Our system demonstrated that even a small set of features for conditional random eld algorithm performed competitively and showed a good results.

The World Wide Web organizes information in semi-structured HTML documents. For a template-based web page that contains a list of items, information schema can be implied and structured data can be extracted with a query, i.e. a (web) wrapper. When a user interface gets updated, the document structure changes, and the wrapper has a tendency to break.

Extracting structured data from multiple data records in a robust way from a single web page is the subject of this project. Current state-of-the-art methods allow to detect template-based data regions on a page, to extract data from a single data record in a robust way, and repeatedly extract data from multiple records.

In this thesis, we have combined the three ideas into a new method, which deals with building a robust record-level wrapper from a single user-annotated web page. We have designed, implemented, and tested the algorithm. Experimental results using a large number of web pages from multiple domains show, that the proposed approach works with high precision and within reasonable execution time on commodity hardware.

In this paper, we present our contribution in SemEval2014 ABSA task, some supervised methods for Aspect-Based Sentiment Analysis of restaurant and laptop reviews are proposed, implemented and evaluated. We focus on determining the aspect terms existing in each sentence, finding out their polarities, detecting the categories of the sentence and the polarity of each category. The evaluation results of our proposed methods exhibit a significant improvement in terms of accuracy and f-measure over all four subtasks regarding to the baseline proposed by SemEval organisers.

This paper introduces a distinctive approach to methods development in digital social research called “interface methods.” We begin by discussing various methodological confluences between digital media, social studies of science and technology (STS) and sociology. Some authors have posited significant overlap between, on the one hand, sociological and STS concepts, and on the other hand, the ontologies of digital media. Others have emphasised the significant differences between prominent methods built into digital media and those of STS and Sociology. This paper advocates a third approach, one that a) highlights the dynamism and relative under-determinacy of digital methods, and b) affirms that multiple methodological traditions intersect in digital devices and research. We argue that these two circumstances enable a distinctive approach to methodology in digital social research - ‘interface methods’ - and the paper contextualizes this approach in two different ways: first, we show how the proliferation of online data tools or ‘digital analytics’ opens up distinctive opportunities for critical and creative engagement with methods development at the intersection of sociology, STS and digital research. Second, we discuss a digital research project in which we investigated a specific ‘interface method’, namely co-occurrence analysis. The second half of the paper presents a digital pilot study in which we implemented this method in a critical and creative way to analyse and visualise ‘issue dynamics’ in the area of climate change on Twitter. We evaluate this project in the light of our principal objective, which was to test the possibilities for the critical and creative adaptation and modification of this method through its experimental implementation. To conclude, we discuss a major obstacle to the development of ‘interface methods’: digital media are marked by particular quantitative dynamics that seem adverse to the methodological commitments of sociology and STS. To address this, we argue in favour of a methodological approach in digital social research that affirms its mal-adjustment to the social methods that are prevalent in the medium.

"This thesis aims to contribute for the discussions on online research methods, by suggesting the concept of a holistic approach to the study of social media. This idea argues that data, online platforms and tools cannot be perceived as isolated objects, but as assemblages of heterogeneous agents. Employing this view, I build a case on research with data from the microblogging site Twitter, analysing the processes of knowledge-making afforded by the platform, the APIs, the database and the software tool for data visualisation Gephi. As a result of this empirical exploration, I argue that social interactions are inherently implicit and appropriated by Twitter users and thus, scholars need to develop tactics that would allow them to look beyond the logic of the database. Moreover, I show that social media research takes place on several levels and on each stage, the method of research is shaped by the actors. Thus, I argue that the concept of online research method cannot be confined within the traditional frames, but instead, it should be perceived as an evolving process, being in a state of constant flux, where heterogeneous actors influence the research decisions, mobilise traditional and new methods and negotiate the research choices of scientific exploration. The main argument of this thesis is that social media studies are not only a technological accomplishment, but should be understood as a complex holistic process.

Key words: online research methods, holistic approach, Twitter, big data, actor-network theory, software tools."

Recently, Savage and Burrows argued that there is an ‘empirical crisis’ in sociology. They concluded that sociologists should abandon a focus on causality for descriptions that ‘link narrative, numbers, and images’. This article takes up their challenge by using Wordle to depict the changing focus in academic articles on food and eating since 1950. Using this illustrative example, it is argued that their call to abandon causality is problematic for three reasons. Firstly, interpreting description necessarily depends on a causal framework. Secondly, since description becomes part of a mode of production in which context and meaning are inscribed, the question is not whether to reject causality in favour of description, but rather what kinds of description help to explore causality. Thirdly, going beyond description is ethically advantageous for a critical sociological programme. The article concludes that, contrary to Savage and Burrows, description and causality go hand in hand.

Background. We are witnessing an exponential increase in biomedical research citations in PubMed. However, translating biomedical discoveries into practical treatments is estimated to take around 17 years, according to the 2000 Yearbook of Medical Informatics, and much information is lost during this transition. Pharmaceutical companies spend huge sums to identify opinion leaders and centers of excellence. Conventional methods such as literature search, survey, observation, self-identification, expert opinion, and sociometry not only need much human effort, but are also noncomprehensive. Such huge delays and costs can be reduced by “connecting those who produce the knowledge with those who apply it”. A humble step in this direction is large scale discovery of persons and organizations involved in specific areas of research. This can be achieved by automatically extracting and disambiguating author names and affiliation strings retrieved through Medical Subject Heading (MeSH) terms and other keywords associated with articles in PubMed. In this study, we propose NEMO (Normalization Engine for Matching Organizations), a system for extracting organization names from the affiliation strings provided in PubMed abstracts, building a thesaurus (list of synonyms) of organization names, and subsequently normalizing them to a canonical organization name using the thesaurus. Results: We used a parsing process that involves multi-layered rule matching with multiple dictionaries. The normalization process involves clustering based on weighted local sequence alignment metrics to address synonymy at word level, and local learning based on finding connected components to address synonymy. The graphical user interface and java client library of NEMO are available at http://lnxnemo.sourceforge.net .Conclusion: NEMO is developed to associate each biomedical paper and its authors with a unique organization name and the geopolitical location of that organization. This system provides more accurate information about organizations than the raw affiliation strings provided in PubMed abstracts. It can be used for : a) bimodal social network analysis that evaluates the research relationships between individual researchers and their institutions; b) improving author name disambiguation; c) augmenting National Library of Medicine (NLM)’s Medical Articles Record System (MARS) system for correcting errors due to OCR on affiliation strings that are in small fonts; and d) improving PubMed citation indexing strategies (authority control) based on normalized organization name and country.