project:hds_out_of_the_box

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
project:hds_out_of_the_box [2016/07/02 16:15] – [Team] timtomproject:hds_out_of_the_box [2016/07/04 15:12] (current) – [Exploring bibliographic enrichment with OpenRefine] pmau
Line 1: Line 1:
 ===== Historical Dictionary of Switzerland Out of the Box ===== ===== Historical Dictionary of Switzerland Out of the Box =====
  
-The [[http://www.hls-dhs-dss.ch/english.php|Historical Dictionary of Switzerland]] (HDS) is an academic reference work which documents ``the most important topics and objects of Swiss history from prehistory up to the present''+The [[http://www.hls-dhs-dss.ch/english.php|Historical Dictionary of Switzerland]] (HDS) is an academic reference work which documents //the most important topics and objects of Swiss history from prehistory up to the present//
  
-The HDS digital edition comprises about XXXX articles organized in 4 main headword groups: \\+The HDS digital edition comprises about 36.000 articles organized in 4 main headword groups: \\
 - Biographies,\\  - Biographies,\\ 
 - Families, \\ - Families, \\
-- Geographical Entities and \\+- Geographical entities and \\
 - Thematical contributions.  - Thematical contributions. 
  
-Beyond the encyclopaedic description of entities/concepts, each articles contains references to primary and secondary sources which supported authors when writing articles.+Beyond the encyclopaedic description of entities/concepts, each article contains references to primary and secondary sources which supported authors when writing articles.
  
  
Line 17: Line 17:
 We have the following data:\\ We have the following data:\\
  
-[[http://make.opendata.ch/wiki/data:glam_ch#metadata_of_the_historical_dictionary_of_switzerland|metadata information]] about HDS articles Historical Dictionary of Switzerland\\ + [[http://make.opendata.ch/wiki/data:glam_ch#metadata_of_the_historical_dictionary_of_switzerland|metadata information]] about HDS articles Historical Dictionary of Switzerland comprising:\\ 
-[[http://make.opendata.ch/wiki/data:glam_ch#journal_de_geneve_gazette_de_lausanne_1914|Le Temps digital archive]] for the year 1914\\ +  * bibliographic references of HDS articles\\ 
-- bibliographic references of HDS articles\\+  * article titles\\ 
 + [[http://make.opendata.ch/wiki/data:glam_ch#journal_de_geneve_gazette_de_lausanne_1914|Le Temps digital archive]] for the year 1914\\ 
      
 ===== Goals ===== ===== Goals =====
  
-Our projects revolve around **Linking the HDS to external data** and aim at:\\+Our projects revolve around **linking the HDS to external data** and aim at:\\
  
-** 1. Entity Linking towards HDS**+  - **Entity linking towards HDS**\\ The objective is to link named entity mentions discovered in historical Swiss newspapers to their correspondant HDS articles.\\
  
-The objective is to link named entity mentions discovered in historical Swiss newspapers to their correspondant HDS articles.+  - **Exploring reference citation of HDS articles**\\ The objective is to reconcile HDS bibliographic data contained in articles with SwissBib.
  
-** 2. Exploring reference citation of HDS articles**  
  
-The objective is to reconcile HDS bibliographic data with SwissBib. 
  
 +===== Named Entity Recognition =====
  
 +We used web-services to annotate text with named entities:\\
 +- Dandelion\\
 +- Alchemy\\
 +- OpenCalais \\
  
-===== Named Entity Recognition ===== 
  
 +{{:project:screen_shot_2016-07-02_at_12.49.39.png?200|}}
 +
 +Named entity mentions (persons and places) are matched against entity labels of HDS entries and directly linked when only one HDS entry exists.
 +
 +Further developments would includes:\\
 +- handling name variants, e.g. 'W.A. Mozart' or 'Mozart' should match 'Wolfgang Amadeus Mozart' .\\
 +- real disambiguation by comparing the newspaper article context with the HDS article context (a first simple similarity could be tf-idf based)\\
 +- working with a more refined NER output which comprises information about name components (first, middle,last names)\\
 +
 +=== Some statistics ===
 +In the 23.622 articles of the year 1914 in «Le Temps digital archive» we linked 90.603 entities pointing to 1.417 articles of the «Historical Dictionary of Switzerland». 
 +
 +{{:project:to15hds.png?500|}}\\
 +
 +
 +=== Web Interface ===
 +
 +We developed a simple web interface for searching in the corpus and displaying the texts with the links.\\
 +It consists of 3 views:
 +
 +
 +1. Home\\
 +{{:project:01home.png?direct&200|}}\\
 +\\
 +2. Search\\
 +{{:project:02search.png?direct&200|}}\\
 +\\
 +3. Article with links to HDS, Wikipedia and dbpedia\\
 +{{:project:03article.png?direct&200|}}\\
 +\\
 +
 +=== Further works ===
 +Further works would include:\\
 +- evaluate and improve method.\\
 +- apply the method to the Historical Dictionary of Switzerland itself for internal linking.\\
  
  
  
-===== Bibliographic Enrichment =====+===== Bibliographic enrichment =====
  
 We work on the list of references in all articles of the HDS, with three goals: We work on the list of references in all articles of the HDS, with three goals:
-  - Finding all the sources which are cited in the HDS (several sources are cited multiple times). +  - Finding all the sources which are cited in the HDS (several sources are cited multiple times) ; 
-  - Link all the sources with the SwissBib catalog, if possible. +  - Link all the sources with the SwissBib catalog, if possible ; 
-  - Interactively explore the citation networks of the HDS.+  - Interactively explore the citation network of the HDS.
  
-The datasetlists of references in every HDS article:+The dataset comes from the HDS metadata. It contains lists of references in every HDS article:
  
 {{:project:hdsrefs.png?direct&500|}} {{:project:hdsrefs.png?direct&500|}}
Line 83: Line 122:
 </code> </code>
  
-Note that the above expression combines the <PUB> column (accessed through value) and the <AUT> column (containing the author's name).+Note that the above expression combines the <PUB> column (accessed through value) and the <AUT> column (containing the author's name). Replace the reference to the authors' column accordingly if it has a different name in your data file.
  
 Swissbib queries can return Dublin Core, MARC XML or MARC in JSON format. Dublin core is the easiest to manipulate, but unfortunately it does not contain the entirety of the returned record. To access the full record, it is necessary to use either MARC XML or MARC JSON. Swissbib queries can return Dublin Core, MARC XML or MARC in JSON format. Dublin core is the easiest to manipulate, but unfortunately it does not contain the entirety of the returned record. To access the full record, it is necessary to use either MARC XML or MARC JSON.
Line 166: Line 205:
  
 (note that the parentheses around "RERO" have to be removed for the search URL to work). (note that the parentheses around "RERO" have to be removed for the search URL to work).
 +
 +=== Further works ===
 +This is only the first step of a more general work inside the HDS:\\
 + * identify precisely each notice in an article (ID attribute to generate)\\
 + * collect references with a separation by language\\
 + * clean and refine the collected data\\
 + * setup a querying workflow that keeps the ID of the matched target in a reference catalog\\
 + * replace each matching occurence in the HDS article by a reference to an external catalog\\
 +
 ===== Team ===== ===== Team =====
  
 +  * Pierre-Marie Aubertel
 +  * Francesco Beretta
   * Giovanni Colavizza   * Giovanni Colavizza
-  * Jonas Schneider 
   * Maud Ehrmann   * Maud Ehrmann
   * [[https://about.me/timtom|Thomas Guignard]]   * [[https://about.me/timtom|Thomas Guignard]]
-  * Pierre-Marie Aubertel +  * Jonas Schneider 
-  * Francesco Beretta+ 
 + 
 +  
  
      
-{{tag>status:concept needs:dev needs:design needs:data needs:expert}}+{{tag>status:concept needs:dev needs:design needs:data needs:expert glam}}
  • project/hds_out_of_the_box.1467468935.txt.gz
  • Last modified: 2016/07/02 16:15
  • by timtom