Temporally-aware dynamic network analysis: traversing nodegoat graphs

CORE Admin

During the conference 'Graphs and Networks in the fourth dimension – time and temporality as categories of connectedness', jointly organised by the Historical Network Research community and Graphs & Networks in the Humanities, we presented a new nodegoat feature: 'temporally-aware dynamic network analysis'. This new functionality extends the Scope and Chronology Statements.

The Scope functionality is used throughout nodegoat to traverse your data model and select elements to be included in a visualisation, analysis, or export. With the Scope, you can limit or expand your data selection. In a prosopographical analyses you might want to include all educational institutes related to one person, plus all the relations of these institutes, while omitting all other personal relations of a person. Follow this Guide to learn how to configure a Scope.

Chronology Statements that you make in nodegoat allow you to specify what you mean by a statement like 'circa'. Instead of using qualitative statements about vagueness, Chronology Statements provide you with a way of making quantitative statements about vagueness. Chronology Statements also allow you to make relational date statements: 'the date point is between the sending of letter X and the sending of letter Y'. Follow this Guide to learn how to store uncertain dates by using Chronology Statements and follow this Guide to learn how to store relational dates by using Chronology Statements.

The temporally-aware dynamic network analysis functionality makes the temporal options offered by the Chronology Statements available on any level of a Scope. This allows you to apply and pass temporality to time-bound connections in any of a Scope's paths. The dates from Chronology Statements can be sourced from every step in the traversal: ascendant or descendant nodes, and combinations. Selected configurations can be applied on any/all of the connections/edges: outbound or inbound directionality, and combinations.

Example: Academic Connections

With this functionality it is now possible to dynamically generate networks of people who attended the same educational institute at the same time, without specifying any dates in a filter. The temporally-aware dynamic network analysis functionality applies the initial date on every other relationship that appears on a specified path:

Two persons shown having an overlapping academic connection out of four persons.

The obvious benefit of this approach is the scalability of this functionality, as it allows you to quickly scrutinise complex networks based on time-bound connections:[....]

Continue readingComment

How to store uncertain data in nodegoat: ambiguous identities

CORE Admin

This blog post is part of a series on storing uncertain datat in nodegoat: 'How to store uncertain data in nodegoat', 'Incomplete source material', 'Conflicting information', 'Ambiguous identities'.

There are many entities that share a name. This is often the case for cities (e.g. Springfield), or people (e.g. Francis Bacon). When you encounter such a name in a source, the context usually provides you with enough clues to know which of the entities is meant. However, in some cases the context is too vague or the entities too similar to be certain. In these cases you need to resort to interpretation and disambiguation. This is genuine scholarly work, since you always have to interpret your sources.

This blog post will describe a case in which disambiguation is needed. We will use the example of a research process that aims to reconstruct scholarly networks in the 17th and 18th century. In a research process that deals with scholarly networks, the source material will largely consist of citations and mentions in documents.

The disambiguation process will be described by means of a snippet taken from a publication by an anonymous author in 1714 with the title 'An account of the Samaritans; in a letter to J---- M------, Esq;' (ESTC Citation No. N16222).

This blog post uses the data model that was created in the nodegoat guide 'Create your first Type', and will use elements from the guide 'Add External Identifiers', and from the guide 'Add Source References'.

To store 'mentioned' statements, you can use the Type that was created in the guide 'Add Source References' and add a new Sub-Object in which mentions can be saved. To change the model, go to Model and edit the Type 'Publication'. Switch to the tab 'Sub-Object' and create a new Sub-Object with the name 'Mention'. Set the Date to 'None' and Location to 'None'. In the tab 'Description', click the green 'add' button twice to create three Sub-Object Descriptions. Name the first 'Person', the second 'Page Number', and the third 'Notes'. Set the value type for 'Person' to 'Reference: Type' and select the Type 'Person'. Set the value type for 'Page Number' to 'Integer' and set the value type for 'Notes' to 'Text'.

These settings are not set in stone. Adjust them so that they work for your project.[....]

Continue readingComment

How to store uncertain data in nodegoat

CORE Admin

This blog post is part of a series on storing uncertain data in nodegoat: 'How to store uncertain data in nodegoat', 'Incomplete source material', 'Conflicting information', 'Ambiguous identities'.

Most scholars think about their research material in terms of nuances, vagueness, and uniqueness, whereas data is perceived as binary, strict, and repetitive. However, working with a digital tool does not mean that you can only work with binary oppositions or uncontested timestamps. On the contrary: by creating a good data model, you are able to include nuances, irregularities, contradictions, and vagueness in your database. A good data model is capable of making these insights and observations explicit. Instead of smoothing out irregularities in the data by simplifying the data model, the model should be adjusted to reflect the existing vagueness, conflicts, and ambiguities.

Before you start to adjust your data model to accommodate uncertainty, you should first try to determine the causes for uncertainty in your data. Most forms of uncertainty in data can be grouped in three categories: incomplete source material, conflicting information, or ambiguous identities.

These types of uncertainty can be dealt with in different ways. The next three blog posts will walk you through a number of possible solutions. The described strategies are not the only possible solutions: each research question is unique and may call for a solution of its own.

Incomplete source material

When the information you need is not available, incomplete, or vague you have to decide if you want to leave the respective parts in your data empty or enter data based on inference or conjecture. Read the blog post 'How to store uncertain data in nodegoat: incomplete source material' to learn how to deal with incomplete source material.

Conflicting information

You might encounter conflicting source material. Two sources might differ about the name of a person, or the date of an event. To account for all possible perspectives, you can include the conflicting statements in your data. Read the blog post 'How to store uncertain data in nodegoat: conflicting information' to learn how to deal with conflicting information.[....]

Continue readingComment

Network Visualisations of 38.000 Letters of 19th Century Intellectuals

CORE Admin

Every bit of information that is entered into nodegoat can immediately be published through a public user interface. This allows the Encyclopedia of Romantic Nationalism in Europe to instantly publish articles and a wide range of research data. This data also includes a set of over 38.000 letters that can be queried through the public user interface. In this blogpost we discuss the steps we took to allow visitors to dynamically explore this dataset.

The Study Platform on Interlocking Nationalisms (SPIN) at the University of Amsterdam has created a dataset of metadata of over 38.000 letters of nineteenth century intellectuals. This data has been manually entered and imported semi-automatically (geo-referencing and disambiguating people was largely done by hand). Sources include a range of publications of letters, like Breve fra og til Carl Christian Rafn, med en biographi, plus two existing datasets: (1) the metadata of over 18.000 letters of Jacob and Wilhelm Grimm were provided by the Arbeitsstelle Grimm-Briefwechsel Berlin, and (2) the metadata of over 14.000 letters of Sir Walter Scott were provided by the Millgate Union Catalogue of W. Scott Correspondence; courtesy prof. Millgate & National Library of Scotland. The remaining 6.000 letters were entered by hand by SPIN, based on publications of letters of various other intellectuals throughout Europe. This means that the dataset is a combination of a number of personal networks and that we have an overrepresentation of letters sent by the people at the center of these personal networks.

This dataset is part of the Encyclopedia of Romantic Nationalism in Europe (ERNiE). ERNiE will include over 1.500 articles on topics and people associated with the era of romantic nationalism (e.g. Dress, design : Romanian, Karadžić, Vuk Stefanović, Felicia Hemans). ERNiE also includes other materials like monuments, architecture, art, and currency. ERNiE is coordinated by SPIN. The editor of ERNiE is Joep Leerssen.[....]

Continue readingComment

Upcoming nodegoat workshops in Ghent & Washington D.C. (and more)

CORE Admin

Next week there will be a nodegoat workshop at the 'DARIAH-EU Annual Meeting' in Gent. This event will take place on 10-13 October. The nodegoat workshop will be on Tuesday 11 October from 14:00 to 15:30. You can find the full program here.

There will also be a nodegoat workshop at the conference 'Creating Spatial Historical Knowledge. New Approaches, Opportunities and Epistemological Implications of Mapping History Digitally'. This conference is organised by the German Historical Institute in Washington DC. The conference takes place on 20-22 October. The nodegoat workshop will be on Thursday 20 October from 14.15 to 16.00. This workshop requires individual registration. The full program of the conference can be found here.

We have proposed a session at the THATCamp Amsterdam on Linked Data challenges. Together with Ingeborg van Vugt we plan to discuss the benefits and difficulties of Linked Data in the humanities.

After a stimulating Virtual Heritage Network conference last year in Maynooth, we look forward to this year's conference in Cork. The conference will take place on 8-10 December.[....]

Continue readingComment

Members of the US House of Representatives - Wikidata

CORE Admin

The following interactive visualisation explores the movements of 10.896 Representatives of the United States Congress, from Roger Sherman's birth in 1721 up until all its members in 2015. The Representatives move from their place of birth to their place of education and finally to their possible place of death. Click here to open the interactive visualisation.

Last April, we gave a talk at the tenth Historical Network Research workshop in Düsseldorf about the 'Reversed Classification' functionality in nodegoat. To illustrate what you can accomplish with this functionality, we queried Wikidata to get a dataset of all the members of the US House of Representatives, including their date and place of birth and death, their professions, and the institutes where they took their education. We used this data to perform a reversed classification process that groups the representatives into career politicians or politicians with a heterogeneous career. From there, you could start looking at geographical patterns or educational backgrounds of these groups. See a graph of this network with these two 'career' nodes included here (canvas).

The diachronic geographical visualisation of all this data in nodegoat turns out to be a nice bonus.

Continue readingComment

nodegoat Workshop in Düsseldorf 28-04-2016

CORE Admin
Düsseldorf, Assumulator / CC BY-SA 3.0

The tenth Historical Network Research workshop will be in Düsseldorf from 28-04-2016 to 30-04-2016. They have set up an exciting programme on the theme 'Fakten verknüpfen, Erkenntnisse gewinnnen? Wissenschaftsgeschichte in Historischer Netzwerkanalyse'.

On the first day, we will host a nodegoat workshop. This workshop will last half a day and is titled 'Advanced HNR' (it will run in parallel with an introductory historical network research workshop by Martin Stark). Since we only have half a day, we encourage participants who have not used nodegoat before to watch our three tutorials that cover basic functionalities of nodegoat.[....]

Continue readingComment

nodegoat as an Interactive Museum Installation: 20.000 letters visualised through time and space

CORE Admin
The installation is located in the first section of the permanent exhibition. The wooden table has a cut-out (elevated) map of Europe as its surface. The visualisation is projected by a Barco F35 projector (WQXGA resolution). Visitors can interact with the installation by means of capacitive sensors.

We have developed an interactive installation for the new GRIMMWELT museum in Kassel, Germany. The installation visualises and lets visitors freely interact with the full correspondence network of Jacob and Wilhelm Grimm, involving a total of 20.000 letters and 1400 correspondence partners in a timespan of 80 years. The dataset of letters has been created by the Arbeitsstelle Grimm-Briefwechsel at the Institut für deutsche Literatur of the  Humboldt-Universität zu Berlin. We have developed the visualisation in cooperation with SPIN: Study Platform on Interlocking Nationalisms at the University of Amsterdam.

The installation implements a new geographical visualisation mode 'Movement' in nodegoat, in addition to the already available line-based 'Connection' mode. The Movement mode uses WebGL rendering (GPU) to animate large collections of objects smoothly. This mode also allows for a wide range of configuration parameters to finetune the visualisation to various scenarios. Due to the open and generic nature of nodegoat, we can now make use of the Movement mode for any other relevant dataset.

This short clip shows the new visualisation mode from within nodegoat:


A high resolution 1440p version of this clip is available here.[....]

Continue readingComment