<h1 id="sentiment-analysis">Sentiment Analysis<a aria-hidden="true" class="anchor-heading icon-link" href="#sentiment-analysis"></a></h1>

<h2 id="significance">Significance<a aria-hidden="true" class="anchor-heading icon-link" href="#significance"></a></h2>
<ul>
<li>What motivated our inquiry?</li>
<li>Who could this research benefit?</li>
<li>How does it relate to the <a href="/notes/7i7dikcobtqreetq3230tqx">broader themes</a>?</li>
</ul>
<h2 id="research-gap">Research Gap<a aria-hidden="true" class="anchor-heading icon-link" href="#research-gap"></a></h2>
<h2 id="opportunity">Opportunity<a aria-hidden="true" class="anchor-heading icon-link" href="#opportunity"></a></h2>
<h3 id="limitations">Limitations<a aria-hidden="true" class="anchor-heading icon-link" href="#limitations"></a></h3>
<ul>
<li>Accessible (open-source, suitable for personal computers) software can perform categorisation based on word frequency (how many 'positive' words appear in this sentence?) but struggles with complex language tasks (detecting sarcasm, negation, qualifications), which can reduce accuracy.</li>
</ul>
<h2 id="methods">Methods<a aria-hidden="true" class="anchor-heading icon-link" href="#methods"></a></h2>
<h2 id="results">Results<a aria-hidden="true" class="anchor-heading icon-link" href="#results"></a></h2>
<ul>
<li>What is the format of the experiment?</li>
<li>How is the experiment accessed?</li>
<li>Represent the experiment
<ul class="contains-task-list">
<li class="task-list-item"><input type="checkbox" disabled> Give more evocative examples</li>
<li class="task-list-item"><input type="checkbox" disabled> Extract .svg sentiment analysis graphs from R as discrete images</li>
</ul>
</li>
</ul>
<h2 id="discussion">Discussion<a aria-hidden="true" class="anchor-heading icon-link" href="#discussion"></a></h2>
<ul>
<li>What research could extend or use these ideas?</li>
</ul>
<hr>
<strong>Backlinks</strong>
<ul>
<li><a href="/notes/z3gbwecwtro4byiji6296gf">Experiments</a></li>
<li><a href="/notes/61wkijq20oig462opkb7nax">Bias in Language</a></li>
<li><a href="/notes/6422iody9ya4ksefq37tn9h">Colour Timeline</a></li>
<li><a href="/notes/tgb8sukw43wku2vizue29as">Text Analysis Overview</a></li>
</ul>

Sentiment Analysis

placecolour

This exhibition explores subjective experiences of place in humans, animals, plants and other living beings. Its objective is to use these sensory ecologies in support of design that benefits all forms of life.

![cover](assets/images/cover.gif)

<sup>Flowers as seen by humans and insects, images by Bjørn Rørslett/NN</sup>



## Structure

![[beginning#^o5fgipezry8p,1:#^p559lhabx8qt]]

## Context

These sections contextualise themes, research questions, methods, and the structure of the exhibition.

### [[Beginning|beginning]]



This section acts as an introduction. It describes the history, intention, components, format, and future of this exhibition.

### [[Ending|ending]]



This section acts as a conclusion. It 1) restates key ideas with examples from the exhibition's [[stories | probes.stories]] and 2) outlines the prospects for future research and creative work.

## Content

These sections organises main exhibition content according to a worldview that sees the world as web of enacted relationships.

### [[What? — Flows|flows]]



The Flows section considers [[actors | flows.portraits]] and  [[relationships | flows.relationships]] as effects of processes. This section provides portraits of phenomena such as colours and sounds,  organisms such as fungi and birds, and communities such as soils and oceans. It also explores relationships that link these phenomena such as cognition, communication, adaptation, and competition.

### [[How? — Probes|probes]]

The Probes applies this process-oriented worldview to examples to question the accepted understandings of the world. We use probes to question preconceptions including the supremacy of complex organisms, the uniqueness and sophistication of human approaches, the desirability of economic growth, the denial of personhood to non-humans, and other related issues.

### [[Why? — Concerns|concerns]]



The Concerns section presents open research questions, study methods, design approaches, and technical recipes that have potential to address ethical issues that result from an inclusive, more-than-human reframing of the world. Examples include use of evidence, measurement, simulation and generative design.




## Authorship

This exhibition is developed by the [Deep Design Lab](https://wiki.deepdesignlab.online/) and collaborators. For more, see the [[acknowledgements | beginning.acknowledgements]] page.

## Usage and Attribution

If you use any of the materials found here, please acknowledge our authorship. If you want to use them for commercial purposes, please get in touch. The contents of this website are licensed under a Creative Commons [Attribution-NonCommercial 4.0](https://creativecommons.org/licenses/by-nc/4.0/) International License (CC BY-NC).

If you would like to get in touch, reach out via [stanislav.roudavski@cantab.net](mailto:stanislav.roudavski@cantab.net)

![Attribution-NonCommercial CC BY-NC](assets/images/license.png){width: 80px}