We are proposing to develop a web-application provisionally called ImpactTracer that allows different types of users to trace the impact of texts on a large number of subsequent texts over time and according to their needs. The collection of texts to be analysed can be assembled by uploading relevant documents or by retrieving them as a result of an internet search. For the process of uploading documents, the application will provide clear instructions as to the accepted formats and data cleansing workflow to follow and how best to achieve this preliminary step. Once the texts are uploaded, the user will then be able to choose what kind of impact s/he wants to measure, which algorithm(s) should be applied accordingly, and how the results should be presented. Presentation of findings could take different forms: firstly, we seek to provide users with quantitative measurements of impact, merging the results of different algorithms and ranking findings according to the accuracy and strength of the underlying method. These can then be viewed online as browsable overviews and summaries or they can be downloaded as a set of reports. Secondly, the goal is to allow for a graphical visualisation of the results, which is particularly important when dealing with a larger collection of documents as quantitative or textual output alone do not allow for an intuitive interpretation of results. The development of impact over time of a group of texts can, for example, be displayed in an interactive timeline with which the user would be able to pan back and forth in time in a visual display of the results.
Call for proposal
See other projects for this call