This shows the gazetteer used to find possible toponym matches in the corpus loaded. The
map interacts with the Context and the Corpus Tools. Features include:
* Zoom in and out buttons; Map View to change to different map styles; and a refresh button.

* Clicking on any toponym in the gazetteer represented as dots in the map, displays further
information about the place. This includes its name, whether this is mentioned in an
historical source, its coordinates, and other information.
* Placename query. Clicking on a toponym in the linked gazetteer represented as dots gives
access to a ‘Search matches’ button. Clicking this updates the Context Tool, showing all
those instances where this toponym is mentioned in the corpus.
* Geographic filter. Several places can be selected using the ‘polygon tool’ in the map. This will
update the Context Tool, showing all those instances where the locations selected are
mentioned in the corpus. To deselect, use the ‘bin button’.
* Clicking on any row in the Context Tool will update the map, showing the location related to
the annotation/keyword/label/phrase in context.
        

The Annotation information box shows further information about the
annotation/term/phrase of interest. This includes the name of the Entity and Label assigned, the
context in which it occurs, and information about any possible match with the geographic
information related to the linked gazetteer.
        

The Context Tool shows the result of the query and its ‘keywords/annotations/labels in
context’. The searched term, phrase, annotation, or label is shown with the context to the
left and right in the sentence. This can be used as an output for further spatial analysis in
Geographic Information Systems or other technologies by using the ‘Export’ function. This
gives the option of exporting the result of the query to .csv (Excel) or .json files. Features
include:
* Clicking on the row to update the Corpus Tool, showing the occurrence in the text.
This will also update the Gazetteer Tool and the Annotation information box
accordingly.
* Further searches filtering the context to the left and right. This can be done typing
any word in the ‘Left context’ ‘Right context’ boxes.
* Adjust the size of the context (i.e. amount of words to be included in the resulting
query to the left and right of the keyword/annotation/labels).
* A count of the matches found in the corpus.
* Export query result to .csv or .json
        

The Corpus tool provides a view of the corpus uploaded. Features include:


* A document index on the left where each document can be turned ‘off’ and ‘on’, and that
can be used to navigate to each document in the ‘Corpus tool’.
* A single click on an unique annotation displays detailed information about that annotation in
the "Annotation information" section. If available, in the Gazetteer Tool, doing a single click
on a unique annotation it also takes you to the place with which that annotation is
associated.
* A double click on an annotation updates de Context Tool, showing each occurrence of such
annotated word with its surrounding text to the left and right (i.e. the context).
* On the upper right there is a switch (‘Text-Tags’) that controls how to search the corpus,
either by double-clicking on a single annotation, or by selecting text with the mouse pointer.

The GTA tool requires a file with the list of entities that are used in the corpus annotation, this file must be in JSON format and with the name "annotations-legend.json".

Below is the structure that the file should carry. This file comes when a tagtog project is downloaded

{
 "e_8" : "Food",
 "e_16" : "Social_Group",
 "e_3" : "Institution",
 "f_28" : "Health",
 "f_24" : "Animal",
 "e_2" : "Date",
 "f_31" : "Date",
 "f_20" : "Gender",
 "e_12" : "Health",
 ...
}

• "e_#: entityName" => A field of an entity is always defined with the letter e at the beginning of the code.
• "f_#: subEntityName" => A field of a sub entity is always defined with the letter f at the beginning of the code. That's fields don't gonna display them in this tool but are necessary.

The gta tool requires document and annotation file sets. When working on a tagtog project, files like this one or the annotation one are created, one of each per worked document.

The documents that contain the text (html file) to be displayed can have two differents structures. The first accepted structure of the file is displayed:

  <!DOCTYPE html >
  <html id="tagtogId-document_title.txt" data-origid="document_title.txt" class="anndoc" data-anndoc-version="3.6" lang="" xml:lang="" xmlns="http://www.w3.org/1999/xhtml">
    <head>
      <meta charset="UTF-8"/>
      <meta name="generator" content="net.tagtog.anndoc.v3.parsers.general.VerbatimParser_v1_0_1"/>
      <title>tagtogId-document_title.txt</title>
    </head>
    <body>
      <article>
        <section data-type="">
          <div class="content">
            <pre id="s1v1">Document content</pre>
          </div>
        </section>
      </article>
    </body>
  </html>
         

The second accepted structure of the file is displayed:

  <!DOCTYPE html >
  <html id="tagtogId-document_title.txt" data-origid="document_title.txt" class="anndoc" data-anndoc-version="3.6" lang="" xml:lang="" xmlns="http://www.w3.org/1999/xhtml">
    <head>
      <meta charset="UTF-8"/>
      <meta name="generator" content="net.tagtog.anndoc.v3.parsers.general.VerbatimParser_v1_0_1"/>
      <title>tagtogId-document_title.txt</title>
    </head>
    <body>
      <article>
        <section data-type="">
          <div class="content">
            <p id="s1p1">Paragraph content</p>
            <p id="s1p2">Paragraph content</p>
            <p id="s1p3">Paragraph content</p>
          </div>
        </section>
      </article>
    </body>
  </html>
         

And finally, for the annotation files they must be in the JSON format with the following structure:

    {
      "annotatable": {
        "parts": ["s1v1"]
      },
      "anncomplete": true,
      "sources": [],
      "metas": {},
      "entities": [
      {
        "classId": "e_4",
        "part": "s1v1",
        "offsets": [{
          "start": 12,
          "text": "toponymName"
        }],
        "coordinates": [],
        "confidence": {
          "state": "pre-added",
          "who": ["user:tagtogUsername"],
          "prob": 1
        },
        "fields": {
          "f_23": {
            "value": "toponym",
            "confidence": {
              "state": "pre-added",
              "who": ["user:tagtogUsername"],
              "prob": 1
            }
          }
        },
        "normalizations": {}
      },
      {
        "classId": "e_4",
        "part": "s1v1",
        "offsets": [{
          "start": 132,
          "text": "toponymName"
        }],
        "coordinates": [],
        "confidence": {
          "state": "pre-added",
          "who": ["user:tagtogUsername"],
          "prob": 1
        },
        "fields": {
          "f_23": {
            "value": "toponym",
            "confidence": {
              "state": "pre-added",
              "who": ["user:tagtogUsername"],
              "prob": 1
            }
          }
        },
        "normalizations": {}
      },
      ...
      ],
      "relations": []
    }
         

To build queries, as a first step, you must select the entity that you want to filter, then the labels that exist for that entity will be shown.

Only one query can be added per entity, the complexity of it will depend on you.

Some sample queries will be shown below: