| { | |
| "paper_id": "S10-1034", | |
| "header": { | |
| "generated_with": "S2ORC 1.0.0", | |
| "date_generated": "2023-01-19T15:28:00.888599Z" | |
| }, | |
| "title": "Likey: Unsupervised Language-independent Keyphrase Extraction", | |
| "authors": [ | |
| { | |
| "first": "Mari-Sanna", | |
| "middle": [], | |
| "last": "Paukkeri", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "Aalto University School of Science and Technology", | |
| "location": { | |
| "postBox": "P.O. Box 15400", | |
| "postCode": "FI-00076", | |
| "settlement": "AALTO", | |
| "country": "Finland" | |
| } | |
| }, | |
| "email": "[email protected]" | |
| }, | |
| { | |
| "first": "Timo", | |
| "middle": [], | |
| "last": "Honkela", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "Aalto University School of Science and Technology", | |
| "location": { | |
| "postBox": "P.O. Box 15400", | |
| "postCode": "FI-00076", | |
| "settlement": "AALTO", | |
| "country": "Finland" | |
| } | |
| }, | |
| "email": "" | |
| } | |
| ], | |
| "year": "", | |
| "venue": null, | |
| "identifiers": {}, | |
| "abstract": "Likey is an unsupervised statistical approach for keyphrase extraction. The method is language-independent and the only language-dependent component is the reference corpus with which the documents to be analyzed are compared. In this study, we have also used another language-dependent component: an English-specific Porter stemmer as a preprocessing step. In our experiments of keyphrase extraction from scientific articles, the Likey method outperforms both supervised and unsupervised baseline methods.", | |
| "pdf_parse": { | |
| "paper_id": "S10-1034", | |
| "_pdf_hash": "", | |
| "abstract": [ | |
| { | |
| "text": "Likey is an unsupervised statistical approach for keyphrase extraction. The method is language-independent and the only language-dependent component is the reference corpus with which the documents to be analyzed are compared. In this study, we have also used another language-dependent component: an English-specific Porter stemmer as a preprocessing step. In our experiments of keyphrase extraction from scientific articles, the Likey method outperforms both supervised and unsupervised baseline methods.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Abstract", | |
| "sec_num": null | |
| } | |
| ], | |
| "body_text": [ | |
| { | |
| "text": "Keyphrase extraction is a natural language processing task for collecting the main topics of a document into a list of phrases. Keyphrases are supposed to be available in the processed documents themselves, and the aim is to extract these most meaningful words and phrases from the documents. Keyphrase extraction summarises the content of a document as few phrases and thus provides a quick way to find out what the document is about. Keyphrase extraction is a basic text mining procedure that can be used as a ground for other, more sophisticated text analysis methods. Automatically extracted keyphrases may be used to improve the performance of information retrieval, automatic user model generation, document collection clustering and visualisation, summarisation and question-answering, among others.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Introduction", | |
| "sec_num": "1" | |
| }, | |
| { | |
| "text": "This article describes the participation of the Likey method in the Task 5 of the SemEval 2010 challenge, automatic keyphrase extraction from scientific articles (Kim et al., 2010) .", | |
| "cite_spans": [ | |
| { | |
| "start": 162, | |
| "end": 180, | |
| "text": "(Kim et al., 2010)", | |
| "ref_id": "BIBREF5" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Introduction", | |
| "sec_num": "1" | |
| }, | |
| { | |
| "text": "In statistical keyphrase extraction, many variations for term frequency counts have been proposed in the literature including relative frequencies (Damerau, 1993) , collection frequency (Hulth, 2003) , term frequency-inverse document frequency (tfidf) (Salton and Buckley, 1988) , among others. Additional features to frequency that have been experimented are e.g., relative position of the first occurrence of the term (Frank et al., 1999) , importance of the sentence in which the term occurs (HaCohen-Kerner, 2003) , and widely studied part-of-speech tag patterns, e.g. Hulth (2003) . Matsuo and Ishizuka (2004) present keyword extraction method using word co-occurrence statistics.", | |
| "cite_spans": [ | |
| { | |
| "start": 147, | |
| "end": 162, | |
| "text": "(Damerau, 1993)", | |
| "ref_id": "BIBREF1" | |
| }, | |
| { | |
| "start": 186, | |
| "end": 199, | |
| "text": "(Hulth, 2003)", | |
| "ref_id": "BIBREF4" | |
| }, | |
| { | |
| "start": 252, | |
| "end": 278, | |
| "text": "(Salton and Buckley, 1988)", | |
| "ref_id": "BIBREF10" | |
| }, | |
| { | |
| "start": 420, | |
| "end": 440, | |
| "text": "(Frank et al., 1999)", | |
| "ref_id": "BIBREF2" | |
| }, | |
| { | |
| "start": 495, | |
| "end": 517, | |
| "text": "(HaCohen-Kerner, 2003)", | |
| "ref_id": "BIBREF3" | |
| }, | |
| { | |
| "start": 573, | |
| "end": 585, | |
| "text": "Hulth (2003)", | |
| "ref_id": "BIBREF4" | |
| }, | |
| { | |
| "start": 588, | |
| "end": 614, | |
| "text": "Matsuo and Ishizuka (2004)", | |
| "ref_id": "BIBREF8" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Related work", | |
| "sec_num": "1.1" | |
| }, | |
| { | |
| "text": "An unsupervised keyphrase extraction method by Liu et al. (2009) uses clustering to find exemplar terms that are then used for keyphrase extraction. Most of the presented methods require a reference corpus or a training corpus to produce keyphrases. Statistical keyphrase extraction methods without reference corpora have also been proposed, e.g. (Matsuo and Ishizuka, 2004; Bracewell et al., 2005) . The later study is carried out for bilingual corpus.", | |
| "cite_spans": [ | |
| { | |
| "start": 47, | |
| "end": 64, | |
| "text": "Liu et al. (2009)", | |
| "ref_id": "BIBREF7" | |
| }, | |
| { | |
| "start": 347, | |
| "end": 374, | |
| "text": "(Matsuo and Ishizuka, 2004;", | |
| "ref_id": "BIBREF8" | |
| }, | |
| { | |
| "start": 375, | |
| "end": 398, | |
| "text": "Bracewell et al., 2005)", | |
| "ref_id": "BIBREF0" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Related work", | |
| "sec_num": "1.1" | |
| }, | |
| { | |
| "text": "The data used in this work are from the SemEval 2010 challenge Task 5, automatic keyphrase extraction from scientific articles. The data consist of train, trial, and test data sets. The number of scientific articles and the total number of word tokens in each of the original data sets (before preprocessing) are given in Table 1 . Three sets of \"correct\" keyphrases are provided for each article in each data set: readerassigned keyphrases, author-provided keyphrases, and a combination of them. All reader-assigned keyphrases have been extracted manually from the papers whereas some of author-provided More detailed information on the data set can be found in (Kim et al., 2010) .", | |
| "cite_spans": [ | |
| { | |
| "start": 663, | |
| "end": 681, | |
| "text": "(Kim et al., 2010)", | |
| "ref_id": "BIBREF5" | |
| } | |
| ], | |
| "ref_spans": [ | |
| { | |
| "start": 322, | |
| "end": 329, | |
| "text": "Table 1", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "Data", | |
| "sec_num": "2" | |
| }, | |
| { | |
| "text": "Likey keyphrase extraction approach comes from the tradition of statistical machine learning (Paukkeri et al., 2008) . The method has been developed to be as language-independent as possible. The only language-specific component needed is a corpus in each language. This kind of data is readily available online or from other sources.", | |
| "cite_spans": [ | |
| { | |
| "start": 93, | |
| "end": 116, | |
| "text": "(Paukkeri et al., 2008)", | |
| "ref_id": "BIBREF9" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Methods", | |
| "sec_num": "3" | |
| }, | |
| { | |
| "text": "Likey selects the words and phrases that best crystallize the meaning of the documents by comparing ranks of frequencies in the documents to those in the reference corpus. The Likey ratio (Paukkeri et al., 2008) for each phrase is defined as", | |
| "cite_spans": [ | |
| { | |
| "start": 188, | |
| "end": 211, | |
| "text": "(Paukkeri et al., 2008)", | |
| "ref_id": "BIBREF9" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Methods", | |
| "sec_num": "3" | |
| }, | |
| { | |
| "text": "EQUATION", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [ | |
| { | |
| "start": 0, | |
| "end": 8, | |
| "text": "EQUATION", | |
| "ref_id": "EQREF", | |
| "raw_str": "L(p, d) = rank d (p) rank r (p) ,", | |
| "eq_num": "(1)" | |
| } | |
| ], | |
| "section": "Methods", | |
| "sec_num": "3" | |
| }, | |
| { | |
| "text": "where rank d (p) is the rank value of phrase p in document d and rank r (p) is the rank value of phrase p in the reference corpus. The rank values are calculated according to the frequencies of phrases of the same length n. If the phrase p does not exist in the reference corpus, the value of the maximum rank for phrases of length n is used: rank r (p) = max rank r (n) + 1. The Likey ratio orders the phrases in a document in such a way that the phrases that have the smallest ratio are the best candidates for being a keyphrase.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Methods", | |
| "sec_num": "3" | |
| }, | |
| { | |
| "text": "As a post-processing step, the phrases of length n > 1 face an extra removal process: if one of the words composing the phrase has a rank of less than a threshold \u03be in the reference corpus, the phrase is removed from the keyphrase list. This procedure excludes phrases that contain function words such as \"of\" or \"the\". As another postprocessing step, phrases that are subphrases of those that have occurred earlier on the keyphrase list are removed, excluding e.g. \"language model\" if \"unigram language model\" has been already accepted as a keyphrase.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Methods", | |
| "sec_num": "3" | |
| }, | |
| { | |
| "text": "Likey needs a reference corpus that is seen as a sample of the general language. In the present study, we use a combination of the English part of Europarl, European Parliament plenary speeches (Koehn, 2005) and the preprocessed training set as the reference corpus. All XML tags of meta information are excluded from the Europarl data. The size of the Europarl corpus is 35 800 000 words after removal of XML tags.", | |
| "cite_spans": [ | |
| { | |
| "start": 194, | |
| "end": 207, | |
| "text": "(Koehn, 2005)", | |
| "ref_id": "BIBREF6" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Reference corpus", | |
| "sec_num": "3.1" | |
| }, | |
| { | |
| "text": "The scientific articles are preprocessed by removing all headers including the names and addresses of the authors. Also the reference section is removed from the articles, as well as all tables, figures, equations and citations. Both scientific articles and the Europarl data is lowercased, punctuation is removed (the hyphens surrounded by word characters and apostrophes are kept) and the numbers are changed to <NUM> tag.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Preprocessing", | |
| "sec_num": "3.2" | |
| }, | |
| { | |
| "text": "The data is stemmed with English Porter stemmer implementation provided by the challenge organizers, which differs from our earlier experiments.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Preprocessing", | |
| "sec_num": "3.2" | |
| }, | |
| { | |
| "text": "We use three baseline methods for keyphrase extraction. The baselines use uni-, bi-, and trigrams as candidates of keyphrases with tf-idf weighting scheme. One of the baselines is unsupervised and the other two are supervised approaches. The unsupervised method is to rank the candidates according to their tf-idf scores. The supervised methods are Na\u00efve Bayes (NB) and Maximum Entropy (ME) implementations from WEKA package 1 .", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Baselines", | |
| "sec_num": "3.3" | |
| }, | |
| { | |
| "text": "We participated the challenge with Likey results of three different parameter settings. The settings are given in Table 3 . Likey-1 has phrases up to 3 words and Likey-2 and Likey-3 up to 4 words. The threshold value for postprocessing was selected against the trial set, with \u03be = 100 performing best. It is used for Likey-1 and Likey-2. Also a bit larger threshold \u03be = 130 was tried for Likey-3 to exclude more function words.", | |
| "cite_spans": [], | |
| "ref_spans": [ | |
| { | |
| "start": 114, | |
| "end": 121, | |
| "text": "Table 3", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "Experiments", | |
| "sec_num": "4" | |
| }, | |
| { | |
| "text": "n \u03be Likey-1 1-3 100 Likey-2 1-4 100 Likey-3 1-4 130 Table 3 : Different parametrizations for Likey: ngram length and threshold value \u03be.", | |
| "cite_spans": [], | |
| "ref_spans": [ | |
| { | |
| "start": 52, | |
| "end": 59, | |
| "text": "Table 3", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "Repr.", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "An example of the resulting keyphrases extracted by Likey-1 from the first scientific article in the test set (article C-1) is given in Table 4. Also the corresponding \"correct\" answers in reader-assigned and author-provided answer sets are shown. The keyphrases are given in stemmed versions. Likey keyphrases that can be found in the reader or author answer sets are emphasized.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Repr.", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Likey-1 uddi registri, proxi registri, servic discoveri, grid servic discoveri, uddi kei, uniqu uddi kei, servic discoveri mechan, distribut hash tabl, web servic, dht, servic name, web servic discoveri, local proxi registri, local uddi registri, queri multipl registri Reader grid servic discoveri, uddi, distribut web-servic discoveri architectur, dht base uddi registri hierarchi, deploy issu, bamboo dht code, case-insensit search, queri, longest avail prefix, qo-base servic discoveri, autonom control, uddi registri, scalabl issu, soft state Author uddi, dht, web servic, grid comput, md, discoveri Table 4 : Extracted keyphrases by Likey-1 from article C-1 and the corresponding correct answers in reader and author answer sets.", | |
| "cite_spans": [], | |
| "ref_spans": [ | |
| { | |
| "start": 605, | |
| "end": 612, | |
| "text": "Table 4", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "Repr.", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The example shows clearly that many of the extracted keyphrases contain the same words that can be found in the correct answer sets but the length of the phrases vary and thus they cannot be counted as successfully extracted keyphrases.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Repr.", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The results for the three different Likey parametrizations and the three baselines are given in Table 5 for reader-assigned keyphrases and Table 6 for the combined set of reader and authorassigned keyphrases. The evaluation is conducted by calculating precision (P), recall (R) and Fmeasure (F) for top 5, 10, and 15 keyphrase candidates for each method, using the reader-assigned and author-provided lists as correct answers. The baseline methods are unsupervised tf-idf and supervised Na\u00efve Bayes (NB) and Maximum Entropy (ME) .", | |
| "cite_spans": [ | |
| { | |
| "start": 499, | |
| "end": 503, | |
| "text": "(NB)", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 524, | |
| "end": 528, | |
| "text": "(ME)", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [ | |
| { | |
| "start": 96, | |
| "end": 103, | |
| "text": "Table 5", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "Repr.", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Likey-1 performed best in the competition and is thus selected as the official result of Likey in the task. Anyway, all Likey parametrizations outperform the baselines, Likey-1 having the best precision 24.60% for top-5 candidates in the reader data set and 29.20% for top-5 candidates in the combined data set. The best F-measure is obtained with Likey-1 for top-10 candidates for both reader and combined data set: 16.24% and 17.11%, respectively. Likey seems to produce the best keyphrases in the beginning of the keyphrase list: for reader-assigned keyphrases the top 5 keyphrase precision for Likey-1 is 6.8 points better than the best-performing baseline tf-idf and the corresponding F-measure is 4.0 points better. For the combined set, the numbers are 7.2 and 3.7 points, respectively. The difference decreases for the larger keyphrase sets.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Repr.", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "This article describes our submission to SemEval 2010 Task 5, keyphrase extraction from scientific articles. Our unsupervised and languageindependent method Likey uses reference corpus and is able to outperform both the unsupervised and supervised baseline methods. The best results are obtained with the top-5 keyphrases: precision of 24.60% with reader-assigned keyphrases and 29.20% with the combination of reader-assigned and author-provided keyphrases.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Conclusions and discussion", | |
| "sec_num": "5" | |
| }, | |
| { | |
| "text": "There are some keyphrases in the answer sets that our method does not find: due to the comparatively large threshold value \u03be many phrases that contain function words, e.g. \"of\", cannot be found. We also extract keyphrases of maximum length of three or four words and thus cannot find keyphrases longer than that. The next step of this research would be to take these problems into account. Table 5 : Results for Likey and the baselines for the reader data set. The best precision (P), recall (R) and F-measure (F) are highlighted. Table 6 : Results for Likey and the baselines for the combined (reader+author) data set. The best precision (P), recall (R) and F-measure (F) are highlighted.", | |
| "cite_spans": [], | |
| "ref_spans": [ | |
| { | |
| "start": 390, | |
| "end": 397, | |
| "text": "Table 5", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 531, | |
| "end": 538, | |
| "text": "Table 6", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "Conclusions and discussion", | |
| "sec_num": "5" | |
| }, | |
| { | |
| "text": "http://www.cs.waikato.ac.nz/\u02dcml/weka/", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "", | |
| "sec_num": null | |
| } | |
| ], | |
| "back_matter": [ | |
| { | |
| "text": "This work was supported by the Finnish Graduate School in Language Studies (Langnet) funded by Ministry of Education of Finland.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Acknowledgements", | |
| "sec_num": null | |
| } | |
| ], | |
| "bib_entries": { | |
| "BIBREF0": { | |
| "ref_id": "b0", | |
| "title": "Multilingual single document keyword extraction for information retrieval", | |
| "authors": [ | |
| { | |
| "first": "David", | |
| "middle": [ | |
| "B" | |
| ], | |
| "last": "Bracewell", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Fuji", | |
| "middle": [], | |
| "last": "Ren", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Shingo", | |
| "middle": [], | |
| "last": "Kuriowa", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2005, | |
| "venue": "Proceedings of NLP-KE'05", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "David B. Bracewell, Fuji Ren, and Shingo Kuriowa. 2005. Multilingual single document keyword ex- traction for information retrieval. In Proceedings of NLP-KE'05.", | |
| "links": null | |
| }, | |
| "BIBREF1": { | |
| "ref_id": "b1", | |
| "title": "Generating and evaluating domain-oriented multi-word terms from text", | |
| "authors": [ | |
| { | |
| "first": "Fred", | |
| "middle": [], | |
| "last": "Damerau", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1993, | |
| "venue": "formation Processing and Management", | |
| "volume": "29", | |
| "issue": "", | |
| "pages": "433--447", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Fred Damerau. 1993. Generating and evaluating domain-oriented multi-word terms from text. In- formation Processing and Management, 29(4):433- 447.", | |
| "links": null | |
| }, | |
| "BIBREF2": { | |
| "ref_id": "b2", | |
| "title": "Domain-specific keyphrase extraction", | |
| "authors": [ | |
| { | |
| "first": "Eibe", | |
| "middle": [], | |
| "last": "Frank", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Gordon", | |
| "middle": [ | |
| "W" | |
| ], | |
| "last": "Paynter", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Ian", | |
| "middle": [ | |
| "H" | |
| ], | |
| "last": "Witten", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Carl", | |
| "middle": [], | |
| "last": "Gutwin", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Craig", | |
| "middle": [ | |
| "G" | |
| ], | |
| "last": "Nevill-Manning", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1999, | |
| "venue": "Proceedings of IJCAI'99", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "668--673", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Eibe Frank, Gordon W. Paynter, Ian H. Witten, Carl Gutwin, and Craig G. Nevill-Manning. 1999. Domain-specific keyphrase extraction. In Proceed- ings of IJCAI'99, pages 668-673.", | |
| "links": null | |
| }, | |
| "BIBREF3": { | |
| "ref_id": "b3", | |
| "title": "Automatic extraction of keywords from abstracts", | |
| "authors": [ | |
| { | |
| "first": "Yaakov", | |
| "middle": [], | |
| "last": "Hacohen-Kerner", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2003, | |
| "venue": "", | |
| "volume": "2773", | |
| "issue": "", | |
| "pages": "843--849", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Yaakov HaCohen-Kerner. 2003. Automatic extrac- tion of keywords from abstracts. In V. Palade, R.J. Howlett, and L.C. Jain, editors, KES 2003, LNAI 2773, pages 843-849. Springer-Verlag.", | |
| "links": null | |
| }, | |
| "BIBREF4": { | |
| "ref_id": "b4", | |
| "title": "Improved automatic keyword extraction given more linguistic knowledge", | |
| "authors": [ | |
| { | |
| "first": "Anette", | |
| "middle": [], | |
| "last": "Hulth", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2003, | |
| "venue": "Proceedings of the Conference on Empirical Methods in Natural Language Processing", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "216--223", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Anette Hulth. 2003. Improved automatic keyword ex- traction given more linguistic knowledge. In Pro- ceedings of the Conference on Empirical Methods in Natural Language Processing, pages 216-223.", | |
| "links": null | |
| }, | |
| "BIBREF5": { | |
| "ref_id": "b5", | |
| "title": "SemEval-2010 Task 5: Automatic Keyphrase Extraction from Scientific Articles", | |
| "authors": [ | |
| { | |
| "first": "Nam", | |
| "middle": [], | |
| "last": "Su", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Alyona", | |
| "middle": [], | |
| "last": "Kim", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Min-Yen", | |
| "middle": [], | |
| "last": "Medelyan", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Timothy", | |
| "middle": [], | |
| "last": "Kan", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "", | |
| "middle": [], | |
| "last": "Baldwin", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2010, | |
| "venue": "Proceedings of the ACL 2010 Workshop on Evaluation Exercises on Semantic Evaluation", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Su Nam Kim, Alyona Medelyan, Min-Yen Kan, and Timothy Baldwin. 2010. SemEval-2010 Task 5: Automatic Keyphrase Extraction from Scientific Ar- ticles. In Proceedings of the ACL 2010 Workshop on Evaluation Exercises on Semantic Evaluation (Se- mEval 2010). to appear.", | |
| "links": null | |
| }, | |
| "BIBREF6": { | |
| "ref_id": "b6", | |
| "title": "Europarl: A parallel corpus for statistical machine translation", | |
| "authors": [ | |
| { | |
| "first": "Philipp", | |
| "middle": [], | |
| "last": "Koehn", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2005, | |
| "venue": "MT Summit", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Philipp Koehn. 2005. Europarl: A parallel corpus for statistical machine translation. In MT Summit 2005.", | |
| "links": null | |
| }, | |
| "BIBREF7": { | |
| "ref_id": "b7", | |
| "title": "Clustering to find exemplar terms for keyphrase extraction", | |
| "authors": [ | |
| { | |
| "first": "Zhiyuan", | |
| "middle": [], | |
| "last": "Liu", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Peng", | |
| "middle": [], | |
| "last": "Li", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Yabin", | |
| "middle": [], | |
| "last": "Zheng", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Maosong", | |
| "middle": [], | |
| "last": "Sun", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2009, | |
| "venue": "Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "257--266", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Zhiyuan Liu, Peng Li, Yabin Zheng, and Maosong Sun. 2009. Clustering to find exemplar terms for keyphrase extraction. In Proceedings of the 2009 Conference on Empirical Methods in Natural Lan- guage Processing, pages 257-266, Singapore, Au- gust. Association for Computational Linguistics.", | |
| "links": null | |
| }, | |
| "BIBREF8": { | |
| "ref_id": "b8", | |
| "title": "Keyword extraction from a single document using word co-occurrence statistical information", | |
| "authors": [ | |
| { | |
| "first": "Yutaka", | |
| "middle": [], | |
| "last": "Matsuo", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Mitsuru", | |
| "middle": [], | |
| "last": "Ishizuka", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2004, | |
| "venue": "International Journal on Artificial Intelligence Tools", | |
| "volume": "13", | |
| "issue": "1", | |
| "pages": "157--169", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Yutaka Matsuo and Mitsuru Ishizuka. 2004. Key- word extraction from a single document using word co-occurrence statistical information. International Journal on Artificial Intelligence Tools, 13(1):157- 169.", | |
| "links": null | |
| }, | |
| "BIBREF9": { | |
| "ref_id": "b9", | |
| "title": "A language-independent approach to keyphrase extraction and evaluation", | |
| "authors": [ | |
| { | |
| "first": "Mari-Sanna", | |
| "middle": [], | |
| "last": "Paukkeri", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "T", | |
| "middle": [], | |
| "last": "Ilari", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Matti", | |
| "middle": [], | |
| "last": "Nieminen", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Timo", | |
| "middle": [], | |
| "last": "P\u00f6ll\u00e4", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "", | |
| "middle": [], | |
| "last": "Honkela", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2008, | |
| "venue": "Coling 2008: Companion volume: Posters", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "83--86", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Mari-Sanna Paukkeri, Ilari T. Nieminen, Matti P\u00f6ll\u00e4, and Timo Honkela. 2008. A language-independent approach to keyphrase extraction and evaluation. In Coling 2008: Companion volume: Posters, pages 83-86, Manchester, UK, August. Coling 2008 Or- ganizing Committee.", | |
| "links": null | |
| }, | |
| "BIBREF10": { | |
| "ref_id": "b10", | |
| "title": "Term weighting approaches in automatic text retrieval. Information Processing and Management", | |
| "authors": [ | |
| { | |
| "first": "Gerard", | |
| "middle": [], | |
| "last": "Salton", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Chris", | |
| "middle": [], | |
| "last": "Buckley", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1988, | |
| "venue": "", | |
| "volume": "24", | |
| "issue": "", | |
| "pages": "513--523", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Gerard Salton and Chris Buckley. 1988. Term weight- ing approaches in automatic text retrieval. Informa- tion Processing and Management, 24(5):513-523.", | |
| "links": null | |
| } | |
| }, | |
| "ref_entries": { | |
| "TABREF0": { | |
| "html": null, | |
| "text": "", | |
| "num": null, | |
| "type_str": "table", | |
| "content": "<table><tr><td>train</td><td>144</td><td colspan=\"2\">1 159 015</td></tr><tr><td>trial</td><td>40</td><td colspan=\"2\">334 379</td></tr><tr><td>test</td><td>100</td><td colspan=\"2\">798 049</td></tr><tr><td colspan=\"4\">Table 1: Number of scientific articles and total</td></tr><tr><td colspan=\"3\">number of word tokens in the data sets.</td><td/></tr><tr><td colspan=\"4\">keyphrases may not occur in the content. The</td></tr><tr><td colspan=\"4\">numbers of correct keyphrases in each data set are</td></tr><tr><td colspan=\"2\">shown in Table 2.</td><td/><td/></tr><tr><td colspan=\"4\">Data set Reader Author Combined</td></tr><tr><td>train</td><td>1 824</td><td>559</td><td>2 223</td></tr><tr><td>trial</td><td>526</td><td>149</td><td>621</td></tr><tr><td>test</td><td>1 204</td><td>387</td><td>1 466</td></tr></table>" | |
| }, | |
| "TABREF1": { | |
| "html": null, | |
| "text": "", | |
| "num": null, | |
| "type_str": "table", | |
| "content": "<table/>" | |
| }, | |
| "TABREF2": { | |
| "html": null, | |
| "text": "Likey-1 24.60 10.22 14.44 17.90 14.87 16.24 13.80 17.19 15.31 Likey-2 23.80 9.88 13.96 16.90 14.04 15.34 13.40 16.69 14.87 Likey-3 23.40 9.72 13.73 16.80 13.95 15.24 13.73 17.11 15.23 tf-idf 17.80 7.39 10.44 13.90 11.54 12.61 11.60 14.45 12.87", | |
| "num": null, | |
| "type_str": "table", | |
| "content": "<table><tr><td>NB</td><td>16.80</td><td>6.98</td><td>9.86 13.30 11.05 12.07 11.40 14.20 12.65</td></tr><tr><td>ME</td><td>16.80</td><td>6.98</td><td>9.86 13.30 11.05 12.07 11.40 14.20 12.65</td></tr></table>" | |
| }, | |
| "TABREF3": { | |
| "html": null, | |
| "text": "Likey-1 29.20 9.96 14.85 21.10 14.39 17.11 16.33 16.71 16.52 Likey-2 28.40 9.69 14.45 19.90 13.57 16.14 15.73 16.10 15.91 Likey-3 28.00 9.55 14.24 19.60 13.37 15.90 16.07 16.44 16.25 tf-idf 22.00 7.50 11.19 17.70 12.07 14.35 14.93 15.28 15.10 NB 21.40 7.30 10.89 17.30 11.80 14.03 14.53 14.87 14.70 ME 21.40 7.30 10.89 17.30 11.80 14.03 14.53 14.87 14.70", | |
| "num": null, | |
| "type_str": "table", | |
| "content": "<table/>" | |
| } | |
| } | |
| } | |
| } |