Skip to content

Latest commit

 

History

History
149 lines (114 loc) · 6.05 KB

README.md

File metadata and controls

149 lines (114 loc) · 6.05 KB
Raw

sideR

sideR has been described in the following manuscripts:

Brief conference paper: Kai Puolamäki, Emilia Oikarinen, Bo Kang, Jefrey Lijffijt, Tijl De Bie. Interactive Visual Data Exploration with Subjective Feedback: An Information-Theoretic Approach. IEEE 34th International Conference on Data Engineering (ICDE), 1208-1211, 2018. https://doi.org/10.1109/ICDE.2018.00112

Extended journal paper: Kai Puolamäki, Emilia Oikarinen, Bo Kang, Jefrey Lijffijt, Tijl De Bie. Interactive visual data exploration with subjective feedback: an information-theoretic approach. Data Min Knowl Disc 34, 21–49 (2020). https://doi.org/10.1007/s10618-019-00655-x

Reference to sideR:

Kai Puolamäki. sideR - a tool for subjective and interactive visual data exploration in R. Downloaded from https://github.com/edahelsinki/sideR on [insert date].

This is the current updated version of the software attached to the above mentioned manuscript.

You need to have the following software installed before running sideR:

We have tested sideR with a unix system (Mac OS X) but it should work in any system that has R with the above mentioned packages installed.

NB: R 3.4.0 includes somes substantial speedups to matrix computations which may affect the performance of the sideR. If you have an older version of R installed we therefore strongly recommended you to update to R 3.4.0 or newer!

Run sideR by executing:
Rscript --vanilla run.R Then point your web browser to http://127.0.0.1:7001.

Run the runtime experiment by executing in directory runtime:
Rscript --vanilla runtime.R
Create the LaTeX table by executing in the same directory:
Rscript --vanilla report.R Produce convergence figure by executing in directory toyexample:
Rscript --vanilla toyexample.R Produce usecases of data exploration and the respective figures by executing in directory usecases: Rscript --vanilla usecases.R

How to get started by sideR:

sideR loads the BNC dataset by default. The BNC dataset contains n=1335 documents which are represented by d=100 dimensional word vectors. You can, e.g., mark a cluster you see on top right hand side with your mouse, painting it red. If you make a mistake you can push "clear current selection". Mark the selection as cluster constraint by pressing "apply cluster constraint to current selection and save". Wait a few seconds and hit "recompute background", after which the background distribution is updated. When the computation is finished you should see how the background distribution changed. You can compute new PCA and ICA projections by pressing "compute pca projection" and "compute ica projection", respectively. You can choose the view (pca/ica/selection) by radio buttons.

You can view the classes and previously saved groupings by the drop-down menu "add to current selection". The status line at the top left corner shows concisely the Jaccard distance between the current selection and clusters. E.g., from the top right cluster should correspond quite well with the texts of the class "Cconversation". Notice that from the axis of the scatterplot you can see the projection vector in terms of the words (since we are here dealing with text documents).

Use cases are described in the manuscript in more detail.

PS. Notice that due in Shiny the ordering of the update of global variables is hard to enforce. This means that sometimes to update a plot you should press "refresh all" button or swap between projections (with radio buttons pca/ica/selection) a couple of time to force the update.

Adding a data set:

You can easily add a data set. You should make a dataset a R data frame with no missing values. The columns that are factors (is.factor(x) true) are automatically converted to class variables visible in the dropdown menu. The other columns are assumed to be real valued attributes. You should save the data file in directory sideR/data and the data file should end with .rds after which it will automatically be read into the dataset dropdown menu when sideR starts.

Data set descriptions and references, see the paper for more discussion:

bnc.rds
Extract from the British National Corpus. BNC 2007. The British National Corpus, version 3 (BNC XML Edition). Distributed by Oxford University Computing Services on behalf of the BNC Consortium. (2007). http://www.natcorp.ox.ac.uk/

segment.rds
UCI Image Segmentation Data Set, https://archive.ics.uci.edu/ml/datasets/Image+Segmentation Lichman, M. (2013). UCI Machine Learning Repository [http://archive.ics.uci.edu/ml]. Irvine, CA: University of California, School of Information and Computer Science.

toy3.rds
toydata.rds
Artificial data sets created by us and described in the paper.

sideR

Copyright (c) 2017-2018 Kai Puolamäki kai.puolamaki@iki.fi
Copyright (c) 2017 Finnish Institute of Occupational Health, Helsinki, Finland

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.