Search (53 results, page 2 of 3)

0.0019955188 = product of:
  0.011973113 = sum of:
    0.011973113 = weight(_text_:in in 3327) [ClassicSimilarity], result of:
      0.011973113 = score(doc=3327,freq=10.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.20163295 = fieldWeight in 3327, product of:
          3.1622777 = tf(freq=10.0), with freq of:
            10.0 = termFreq=10.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=3327)
  0.16666667 = coord(1/6)

Abstract

Social network sites (SNSs) such as MySpace and Facebook are important venues for interpersonal communication, especially among youth. One way in which members can communicate is to write public messages on each other's profile, but how is this unusual means of communication used in practice? An analysis of 2,293 public comment exchanges extracted from large samples of U.S. and U.K. MySpace members found them to be relatively rapid, but rarely used for prolonged exchanges. They seem to fulfill two purposes: making initial contact and keeping in touch occasionally such as at birthdays and other important dates. Although about half of the dialogs seem to exchange some gossip, the dialogs seem typically too short to play the role of gossip-based social grooming for typical pairs of Friends, but close Friends may still communicate extensively in SNSs with other methods.

0.0019955188 = product of:
  0.011973113 = sum of:
    0.011973113 = weight(_text_:in in 521) [ClassicSimilarity], result of:
      0.011973113 = score(doc=521,freq=10.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.20163295 = fieldWeight in 521, product of:
          3.1622777 = tf(freq=10.0), with freq of:
            10.0 = termFreq=10.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=521)
  0.16666667 = coord(1/6)

Abstract

Applying text mining techniques to legal issues has been an emerging research topic in recent years. Although some previous studies focused on assisting professionals in the retrieval of related legal documents, they did not take into account the general public and their difficulty in describing legal problems in professional legal terms. Because this problem has not been addressed by previous research, this study aims to design a text-mining-based method that allows the general public to use everyday vocabulary to search for and retrieve criminal judgments. The experimental results indicate that our method can help the general public, who are not familiar with professional legal terms, to acquire relevant criminal judgments more accurately and effectively.

0.0019955188 = product of:
  0.011973113 = sum of:
    0.011973113 = weight(_text_:in in 605) [ClassicSimilarity], result of:
      0.011973113 = score(doc=605,freq=10.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.20163295 = fieldWeight in 605, product of:
          3.1622777 = tf(freq=10.0), with freq of:
            10.0 = termFreq=10.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=605)
  0.16666667 = coord(1/6)

Abstract

With the development of Web2.0, social tagging systems in which users can freely choose tags to annotate resources according to their interests have attracted much attention. In particular, literature on the emergence of collective intelligence in social tagging systems has increased. In this article, we propose a probabilistic generative model to detect latent topical communities among users. Social tags and resource contents are leveraged to model user interest in two similar and correlated ways. Our primary goal is to capture user tagging behavior and interest and discover the emergent topical community structure. The communities should be groups of users with frequent social interactions as well as similar topical interests, which would have important research implications for personalized information services. Experimental results on two real social tagging data sets with different genres have shown that the proposed generative model more accurately models user interest and detects high-quality and meaningful topical communities.

0.0019676082 = product of:
  0.011805649 = sum of:
    0.011805649 = weight(_text_:in in 1286) [ClassicSimilarity], result of:
      0.011805649 = score(doc=1286,freq=14.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.19881277 = fieldWeight in 1286, product of:
          3.7416575 = tf(freq=14.0), with freq of:
            14.0 = termFreq=14.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=1286)
  0.16666667 = coord(1/6)

Abstract

With the rapid growth of Web 2.0, community question answering (CQA) has become a prevalent information seeking channel, in which users form interactive communities by posting questions and providing answers. Communities may evolve over time, because of changes in users' interests, activities, and new users joining the network. To better understand user interactions in CQA communities, it is necessary to analyze the community structures and track community evolution over time. Existing work in CQA focuses on question searching or content quality detection, and the important problems of community extraction and evolutionary pattern detection have not been studied. In this article, we propose a probabilistic community model (PCM) to extract overlapping community structures and capture their evolution patterns in CQA. The empirical results show that our algorithm appears to improve the community extraction quality. We show empirically, using the iPhone data set, that interesting community evolution patterns can be discovered, with each evolution pattern reflecting the variation of users' interests over time. Our analysis suggests that individual users could benefit to gain comprehensive information from tracking the transition of products. We also show that the communities provide a decision-making basis for business.

0.001821651 = product of:
  0.010929906 = sum of:
    0.010929906 = weight(_text_:in in 1001) [ClassicSimilarity], result of:
      0.010929906 = score(doc=1001,freq=12.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.18406484 = fieldWeight in 1001, product of:
          3.4641016 = tf(freq=12.0), with freq of:
            12.0 = termFreq=12.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=1001)
  0.16666667 = coord(1/6)

Abstract

When it comes to evaluating online information experiences, what metrics matter? We conducted a study in which 30 people browsed and selected content within an online news website. Data collected included psychometric scales (User Engagement, Cognitive Absorption, System Usability Scales), self-reported interest in news content, and performance metrics (i.e., reading time, browsing time, total time, number of pages visited, and use of recommended links); a subset of the participants had their physiological responses recorded during the interaction (i.e., heart rate, electrodermal activity, electrocmytogram). Findings demonstrated the concurrent validity of the psychometric scales and interest ratings and revealed that increased time on tasks, number of pages visited, and use of recommended links were not necessarily indicative of greater self-reported engagement, cognitive absorption, or perceived usability. Positive ratings of news content were associated with lower physiological activity. The implications of this research are twofold. First, we propose that user experience is a useful framework for studying online information interactions and will result in a broader conceptualization of information interaction and its evaluation. Second, we advocate a mixed-methods approach to measurement that employs a suite of metrics capable of capturing the pragmatic (e.g., usability) and hedonic (e.g., fun, engagement) aspects of information interactions. We underscore the importance of using multiple measures in information research, because our results emphasize that performance and physiological data must be interpreted in the context of users' subjective experiences.

0.001821651 = product of:
  0.010929906 = sum of:
    0.010929906 = weight(_text_:in in 2155) [ClassicSimilarity], result of:
      0.010929906 = score(doc=2155,freq=12.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.18406484 = fieldWeight in 2155, product of:
          3.4641016 = tf(freq=12.0), with freq of:
            12.0 = termFreq=12.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=2155)
  0.16666667 = coord(1/6)

Abstract

The recent interest in Big Data has generated a broad range of new academic, corporate, and policy practices along with an evolving debate among its proponents, detractors, and skeptics. While the practices draw on a common set of tools, techniques, and technologies, most contributions to the debate come either from a particular disciplinary perspective or with a focus on a domain-specific issue. A close examination of these contributions reveals a set of common problematics that arise in various guises and in different places. It also demonstrates the need for a critical synthesis of the conceptual and practical dilemmas surrounding Big Data. The purpose of this article is to provide such a synthesis by drawing on relevant writings in the sciences, humanities, policy, and trade literature. In bringing these diverse literatures together, we aim to shed light on the common underlying issues that concern and affect all of these areas. By contextualizing the phenomenon of Big Data within larger socioeconomic developments, we also seek to provide a broader understanding of its drivers, barriers, and challenges. This approach allows us to identify attributes of Big Data that require more attention-autonomy, opacity, generativity, disparity, and futurity-leading to questions and ideas for moving beyond dilemmas.

Series

Advances in information science

0.0018033426 = product of:
  0.010820055 = sum of:
    0.010820055 = weight(_text_:in in 3886) [ClassicSimilarity], result of:
      0.010820055 = score(doc=3886,freq=6.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.1822149 = fieldWeight in 3886, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0546875 = fieldNorm(doc=3886)
  0.16666667 = coord(1/6)

Abstract

The paper investigates the acceleration of t-SNE-an embedding technique that is commonly used for the visualization of high-dimensional data in scatter plots-using two tree-based algorithms. In particular, the paper develops variants of the Barnes-Hut algorithm and of the dual-tree algorithm that approximate the gradient used for learning t-SNE embeddings in O(N*logN). Our experiments show that the resulting algorithms substantially accelerate t-SNE, and that they make it possible to learn embeddings of data sets with millions of objects. Somewhat counterintuitively, the Barnes-Hut variant of t-SNE appears to outperform the dual-tree variant.

0.0017848461 = product of:
  0.010709076 = sum of:
    0.010709076 = weight(_text_:in in 3464) [ClassicSimilarity], result of:
      0.010709076 = score(doc=3464,freq=8.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.18034597 = fieldWeight in 3464, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=3464)
  0.16666667 = coord(1/6)

Abstract

We propose a new hybrid clustering framework to incorporate text mining with bibliometrics in journal set analysis. The framework integrates two different approaches: clustering ensemble and kernel-fusion clustering. To improve the flexibility and the efficiency of processing large-scale data, we propose an information-based weighting scheme to leverage the effect of multiple data sources in hybrid clustering. Three different algorithms are extended by the proposed weighting scheme and they are employed on a large journal set retrieved from the Web of Science (WoS) database. The clustering performance of the proposed algorithms is systematically evaluated using multiple evaluation methods, and they were cross-compared with alternative methods. Experimental results demonstrate that the proposed weighted hybrid clustering strategy is superior to other methods in clustering performance and efficiency. The proposed approach also provides a more refined structural mapping of journal sets, which is useful for monitoring and detecting new trends in different scientific fields.

0.0017848461 = product of:
  0.010709076 = sum of:
    0.010709076 = weight(_text_:in in 1326) [ClassicSimilarity], result of:
      0.010709076 = score(doc=1326,freq=8.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.18034597 = fieldWeight in 1326, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=1326)
  0.16666667 = coord(1/6)

Abstract

The effective use of corporate memory is becoming increasingly important because every aspect of e-business requires access to information repositories. Unfortunately, less-than-satisfying effectiveness in state-of-the-art information-retrieval techniques is well known, even for some of the best search engines such as Google. In this study, the authors resolve this retrieval ineffectiveness problem by developing a new framework for predicting query performance, which is the first step toward better retrieval effectiveness. Specifically, they examine the relationship between query performance and query context. A query context consists of the query itself, the document collection, and the interaction between the two. The authors first analyze the characteristics of query context and develop various features for predicting query performance. Then, they propose a context-sensitive model for predicting query performance based on the characteristics of the query and the document collection. Finally, they validate this model with respect to five real-world collections of documents and demonstrate its utility in routing queries to the correct repository with high accuracy.

0.001682769 = product of:
  0.010096614 = sum of:
    0.010096614 = weight(_text_:in in 505) [ClassicSimilarity], result of:
      0.010096614 = score(doc=505,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.17003182 = fieldWeight in 505, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0625 = fieldNorm(doc=505)
  0.16666667 = coord(1/6)

Abstract

The behavior of modern web robots varies widely when they crawl for different purposes. In this article, we present a framework to classify these web robots from two orthogonal perspectives, namely, their functionality and the types of resources they consume. Applying the classification framework to a year-long access log from the UConn SoE web server, we present trends that point to significant differences in their crawling behavior.

0.0016629322 = product of:
  0.009977593 = sum of:
    0.009977593 = weight(_text_:in in 4367) [ClassicSimilarity], result of:
      0.009977593 = score(doc=4367,freq=10.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.16802745 = fieldWeight in 4367, product of:
          3.1622777 = tf(freq=10.0), with freq of:
            10.0 = termFreq=10.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=4367)
  0.16666667 = coord(1/6)

Abstract

Information extraction is an important text-mining task that aims at extracting prespecified types of information from large text collections and making them available in structured representations such as databases. In the biomedical domain, information extraction can be applied to help biologists make the most use of their digital-literature archives. Currently, there are large amounts of biomedical literature that contain rich information about biomedical substances. Extracting such knowledge requires a good named entity recognition technique. In this article, we combine conditional random fields (CRFs), a state-of-the-art sequence-labeling algorithm, with two semisupervised learning techniques, bootstrapping and feature sampling, to recognize disease names from biomedical literature. Two data-processing strategies for each technique also were analyzed: one sequentially processing unlabeled data partitions and another one processing unlabeled data partitions in a round-robin fashion. The experimental results showed the advantage of semisupervised learning techniques given limited labeled training data. Specifically, CRFs with bootstrapping implemented in sequential fashion outperformed strictly supervised CRFs for disease name recognition. The project was supported by NIH/NLM Grant R33 LM07299-01, 2002-2005.

0.0016629322 = product of:
  0.009977593 = sum of:
    0.009977593 = weight(_text_:in in 3607) [ClassicSimilarity], result of:
      0.009977593 = score(doc=3607,freq=10.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.16802745 = fieldWeight in 3607, product of:
          3.1622777 = tf(freq=10.0), with freq of:
            10.0 = termFreq=10.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3607)
  0.16666667 = coord(1/6)

Abstract

The abundance of medical resources has encouraged the development of systems that allow for efficient searches of information in large medical image data sets. State-of-the-art image retrieval models are classified into three categories: content-based (visual) models, textual models, and combined models. Content-based models use visual features to answer image queries, textual image retrieval models use word matching to answer textual queries, and combined image retrieval models, use both textual and visual features to answer queries. Nevertheless, most of previous works in this field have used the same image retrieval model independently of the query type. In this article, we define a list of generic and specific medical query features and exploit them in an association rule mining technique to discover correlations between query features and image retrieval models. Based on these rules, we propose to use an associative classifier (NaiveClass) to find the best suitable retrieval model given a new textual query. We also propose a second associative classifier (SmartClass) to select the most appropriate default class for the query. Experiments are performed on Medical ImageCLEF queries from 2008 to 2012 to evaluate the impact of the proposed query features on the classification performance. The results show that combining our proposed specific and generic query features is effective in query classification.

0.0016629322 = product of:
  0.009977593 = sum of:
    0.009977593 = weight(_text_:in in 4019) [ClassicSimilarity], result of:
      0.009977593 = score(doc=4019,freq=10.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.16802745 = fieldWeight in 4019, product of:
          3.1622777 = tf(freq=10.0), with freq of:
            10.0 = termFreq=10.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=4019)
  0.16666667 = coord(1/6)

Abstract

What is text mining, and how can it be used? What relevance do these methods have to everyday work in information science and the digital humanities? How does one develop competences in text mining? Working with Text provides a series of cross-disciplinary perspectives on text mining and its applications. As text mining raises legal and ethical issues, the legal background of text mining and the responsibilities of the engineer are discussed in this book. Chapters provide an introduction to the use of the popular GATE text mining package with data drawn from social media, the use of text mining to support semantic search, the development of an authority system to support content tagging, and recent techniques in automatic language evaluation. Focused studies describe text mining on historical texts, automated indexing using constrained vocabularies, and the use of natural language processing to explore the climate science literature. Interviews are included that offer a glimpse into the real-life experience of working within commercial and academic text mining.

Footnote

Rez. in: JASIST 69(2018) no.1, S.181-184 (Jacques Savoy).

0.0015740865 = product of:
  0.009444519 = sum of:
    0.009444519 = weight(_text_:in in 362) [ClassicSimilarity], result of:
      0.009444519 = score(doc=362,freq=14.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.15905021 = fieldWeight in 362, product of:
          3.7416575 = tf(freq=14.0), with freq of:
            14.0 = termFreq=14.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.03125 = fieldNorm(doc=362)
  0.16666667 = coord(1/6)

Abstract

Text mining applications have experienced tremendous advances because of web 2.0 and social networking applications. Recent advances in hardware and software technology have lead to a number of unique scenarios where text mining algorithms are learned. Mining Text Data introduces an important niche in the text analytics field, and is an edited volume contributed by leading international researchers and practitioners focused on social networks & data mining. This book contains a wide swath in topics across social networks & data mining. Each chapter contains a comprehensive survey including the key research content on the topic, and the future directions of research in the field. There is a special focus on Text Embedded with Heterogeneous and Multimedia Data which makes the mining process much more challenging. A number of methods have been designed such as transfer learning and cross-lingual mining for such cases. Mining Text Data simplifies the content, so that advanced-level students, practitioners and researchers in computer science can benefit from this book. Academic and corporate libraries, as well as ACM, IEEE, and Management Science focused on information security, electronic commerce, databases, data mining, machine learning, and statistics are the primary buyers for this reference book.

Content

Inhalt: An Introduction to Text Mining.- Information Extraction from Text.- A Survey of Text Summarization Techniques.- A Survey of Text Clustering Algorithms.- Dimensionality Reduction and Topic Modeling.- A Survey of Text Classification Algorithms.- Transfer Learning for Text Mining.- Probabilistic Models for Text Mining.- Mining Text Streams.- Translingual Mining from Text Data.- Text Mining in Multimedia.- Text Analytics in Social Media.- A Survey of Opinion Mining and Sentiment Analysis.- Biomedical Text Mining: A Survey of Recent Progress.- Index.

0.0015457221 = product of:
  0.009274333 = sum of:
    0.009274333 = weight(_text_:in in 4226) [ClassicSimilarity], result of:
      0.009274333 = score(doc=4226,freq=6.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.1561842 = fieldWeight in 4226, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=4226)
  0.16666667 = coord(1/6)

Abstract

Many powerful methods and tools exist for extracting meaning from scientific publications, their texts, and their citation links. However, existing proposals often neglect a fundamental aspect of learning: that understanding and learning require an active and constructive exploration of a domain. In this paper, we describe a new method and a tool that use data mining and interactivity to turn the typical search and retrieve dialogue, in which the user asks questions and a system gives answers, into a dialogue that also involves sense-making, in which the user has to become active by constructing a bibliography and a domain model of the search term(s). This model starts from an automatically generated and annotated clustering solution that is iteratively modified by users. The tool is part of an integrated authoring system covering all phases from search through reading and sense-making to writing. Two evaluation studies demonstrate the usability of this interactive and constructive approach, and they show that clusters and groups represent identifiable sub-topics.

0.0015457221 = product of:
  0.009274333 = sum of:
    0.009274333 = weight(_text_:in in 3111) [ClassicSimilarity], result of:
      0.009274333 = score(doc=3111,freq=6.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.1561842 = fieldWeight in 3111, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=3111)
  0.16666667 = coord(1/6)

Abstract

Modern organizations often employ data scientists to improve business processes using diverse sets of data. Researchers and practitioners have both touted the benefits and warned of the drawbacks associated with data science and big data approaches, but few studies investigate how data science is carried out "on the ground." In this paper, we first review the hype and criticisms surrounding data science and big data approaches. We then present the findings of semistructured interviews with 18 data analysts from various industries and organizational roles. Using qualitative coding techniques, we evaluated these interviews in light of the hype and criticisms surrounding data science in the popular discourse. We found that although the data analysts we interviewed were sensitive to both the allure and the potential pitfalls of data science, their motivations and evaluations of their work were more nuanced. We conclude by reflecting on the relationship between data analysts' work and the discourses around data science and big data, suggesting how future research can better account for the everyday practices of this profession.

0.0014873719 = product of:
  0.008924231 = sum of:
    0.008924231 = weight(_text_:in in 967) [ClassicSimilarity], result of:
      0.008924231 = score(doc=967,freq=8.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.15028831 = fieldWeight in 967, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=967)
  0.16666667 = coord(1/6)

Abstract

Because of Twitter's popularity and the viral nature of information dissemination on Twitter, predicting which Twitter topics will become popular in the near future becomes a task of considerable economic importance. Many Twitter topics are annotated by hashtags. In this article, we propose methods to predict the popularity of new hashtags on Twitter by formulating the problem as a classification task. We use five standard classification models (i.e., Naïve bayes, k-nearest neighbors, decision trees, support vector machines, and logistic regression) for prediction. The main challenge is the identification of effective features for describing new hashtags. We extract 7 content features from a hashtag string and the collection of tweets containing the hashtag and 11 contextual features from the social graph formed by users who have adopted the hashtag. We conducted experiments on a Twitter data set consisting of 31 million tweets from 2 million Singapore-based users. The experimental results show that the standard classifiers using the extracted features significantly outperform the baseline methods that do not use these features. Among the five classifiers, the logistic regression model performs the best in terms of the Micro-F1 measure. We also observe that contextual features are more effective than content features.

0.0014873719 = product of:
  0.008924231 = sum of:
    0.008924231 = weight(_text_:in in 3540) [ClassicSimilarity], result of:
      0.008924231 = score(doc=3540,freq=8.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.15028831 = fieldWeight in 3540, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3540)
  0.16666667 = coord(1/6)

Abstract

Opinion mining refers to the use of natural language processing, text analysis, and computational linguistics to identify and extract subjective information in textual material. Opinion mining, also known as sentiment analysis, has received a lot of attention in recent times, as it provides a number of tools to analyze public opinion on a number of different topics. Comparative opinion mining is a subfield of opinion mining which deals with identifying and extracting information that is expressed in a comparative form (e.g., "paper X is better than the Y"). Comparative opinion mining plays a very important role when one tries to evaluate something because it provides a reference point for the comparison. This paper provides a review of the area of comparative opinion mining. It is the first review that cover specifically this topic as all previous reviews dealt mostly with general opinion mining. This survey covers comparative opinion mining from two different angles. One from the perspective of techniques and the other from the perspective of comparative opinion elements. It also incorporates preprocessing tools as well as data set that were used by past researchers that can be useful to future researchers in the field of comparative opinion mining.

0.0014873719 = product of:
  0.008924231 = sum of:
    0.008924231 = weight(_text_:in in 3682) [ClassicSimilarity], result of:
      0.008924231 = score(doc=3682,freq=8.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.15028831 = fieldWeight in 3682, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3682)
  0.16666667 = coord(1/6)

Abstract

Analyzing large quantities of real-world textual data has the potential to provide new insights for researchers. However, such data present challenges for both human and computational methods, requiring a diverse range of specialist skills, often shared across a number of individuals. In this paper we use the analysis of a real-world data set as our case study, and use this exploration as a demonstration of our "insight workflow," which we present for use and adaptation by other researchers. The data we use are impact case study documents collected as part of the UK Research Excellence Framework (REF), consisting of 6,679 documents and 6.25 million words; the analysis was commissioned by the Higher Education Funding Council for England (published as report HEFCE 2015). In our exploration and analysis we used a variety of techniques, ranging from keyword in context and frequency information to more sophisticated methods (topic modeling), with these automated techniques providing an empirical point of entry for in-depth and intensive human analysis. We present the 60 topics to demonstrate the output of our methods, and illustrate how the variety of analysis techniques can be combined to provide insights. We note potential limitations and propose future work.

0.0012881019 = product of:
  0.007728611 = sum of:
    0.007728611 = weight(_text_:in in 3059) [ClassicSimilarity], result of:
      0.007728611 = score(doc=3059,freq=6.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.1301535 = fieldWeight in 3059, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3059)
  0.16666667 = coord(1/6)

Abstract

This study proposes a new method to construct and trace the trajectory of conceptual development of a research field by combining main path analysis, citation analysis, and text-mining techniques. Main path analysis, a method used commonly to trace the most critical path in a citation network, helps describe the developmental trajectory of a research field. This study extends the main path analysis method and applies text-mining techniques in the new method, which reflects the trajectory of conceptual development in an academic research field more accurately than citation frequency, which represents only the articles examined. Articles can be merged based on similarity of concepts, and by merging concepts the history of a research field can be described more precisely. The new method was applied to the "h-index" and "text mining" fields. The precision, recall, and F-measures of the h-index were 0.738, 0.652, and 0.658 and those of text-mining were 0.501, 0.653, and 0.551, respectively. Last, this study not only establishes the conceptual trajectory map of a research field, but also recommends keywords that are more precise than those used currently by researchers. These precise keywords could enable researchers to gather related works more quickly than before.