Search (2 results, page 1 of 1)

0.01938208 = product of:
  0.03876416 = sum of:
    0.03876416 = sum of:
      0.0075639198 = weight(_text_:a in 2648) [ClassicSimilarity], result of:
        0.0075639198 = score(doc=2648,freq=10.0), product of:
          0.053105544 = queryWeight, product of:
            1.153047 = idf(docFreq=37942, maxDocs=44218)
            0.046056706 = queryNorm
          0.14243183 = fieldWeight in 2648, product of:
            3.1622777 = tf(freq=10.0), with freq of:
              10.0 = termFreq=10.0
            1.153047 = idf(docFreq=37942, maxDocs=44218)
            0.0390625 = fieldNorm(doc=2648)
      0.03120024 = weight(_text_:22 in 2648) [ClassicSimilarity], result of:
        0.03120024 = score(doc=2648,freq=2.0), product of:
          0.16128273 = queryWeight, product of:
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.046056706 = queryNorm
          0.19345059 = fieldWeight in 2648, product of:
            1.4142135 = tf(freq=2.0), with freq of:
              2.0 = termFreq=2.0
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.0390625 = fieldNorm(doc=2648)
  0.5 = coord(1/2)

Abstract

The growing predominance of social semantics in the form of tagging presents the metadata community with both opportunities and challenges as for leveraging this new form of information content representation and for retrieval. One key challenge is the absence of contextual information associated with these tags. This paper presents an experiment working with Flickr tags as an example of utilizing social semantics sources for enriching subject metadata. The procedure included four steps: 1) Collecting a sample of Flickr tags, 2) Calculating cooccurrences between tags through mutual information, 3) Tracing contextual information of tag pairs via Google search results, 4) Applying natural language processing and machine learning techniques to extract semantic relations between tags. The experiment helped us to build a context sentence collection from the Google search results, which was then processed by natural language processing and machine learning algorithms. This new approach achieved a reasonably good rate of accuracy in assigning semantic relations to tag pairs. This paper also explores the implications of this approach for using social semantics to enrich subject metadata.

Source

Metadata for semantic and social applications : proceedings of the International Conference on Dublin Core and Metadata Applications, Berlin, 22 - 26 September 2008, DC 2008: Berlin, Germany / ed. by Jane Greenberg and Wolfgang Klas

Type

a

0.0023435948 = product of:
  0.0046871896 = sum of:
    0.0046871896 = product of:
      0.009374379 = sum of:
        0.009374379 = weight(_text_:a in 872) [ClassicSimilarity], result of:
          0.009374379 = score(doc=872,freq=24.0), product of:
            0.053105544 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046056706 = queryNorm
            0.17652355 = fieldWeight in 872, product of:
              4.8989797 = tf(freq=24.0), with freq of:
                24.0 = termFreq=24.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.03125 = fieldNorm(doc=872)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Recent work has demonstrated substantial gains on many NLP tasks and benchmarks by pre-training on a large corpus of text followed by fine-tuning on a specific task. While typically task-agnostic in architecture, this method still requires task-specific fine-tuning datasets of thousands or tens of thousands of examples. By contrast, humans can generally perform a new language task from only a few examples or from simple instructions - something which current NLP systems still largely struggle to do. Here we show that scaling up language models greatly improves task-agnostic, few-shot performance, sometimes even reaching competitiveness with prior state-of-the-art fine-tuning approaches. Specifically, we train GPT-3, an autoregressive language model with 175 billion parameters, 10x more than any previous non-sparse language model, and test its performance in the few-shot setting. For all tasks, GPT-3 is applied without any gradient updates or fine-tuning, with tasks and few-shot demonstrations specified purely via text interaction with the model. GPT-3 achieves strong performance on many NLP datasets, including translation, question-answering, and cloze tasks, as well as several tasks that require on-the-fly reasoning or domain adaptation, such as unscrambling words, using a novel word in a sentence, or performing 3-digit arithmetic. At the same time, we also identify some datasets where GPT-3's few-shot learning still struggles, as well as some datasets where GPT-3 faces methodological issues related to training on large web corpora. Finally, we find that GPT-3 can generate samples of news articles which human evaluators have difficulty distinguishing from articles written by humans. We discuss broader societal impacts of this finding and of GPT-3 in general.

Type

a