Updates (#41)

Website copy edits and minor updates.

Author: s2t2

src/_App/BotAnalysis/Beneficiaries/Section.js

@@ -23,9 +23,6 @@ export default function BotBeneficiariesSection() {
                     </Card.Text>
 
                     <h4>Users Most Retweeted by Bot Community</h4>
-                    <Card.Text>
-                        We observe left-leaning bots retweet a greater variety of users, while right-leaning bots retweet Trump significantly more than any other user.
-                    </Card.Text>
                     <UsersMostRetweeted/>
 
                     <h4>Statuses Most Retweeted by Bot Community</h4>

src/_App/BotAnalysis/Beneficiaries/StatusesMostRetweetedTable.js

@@ -25,7 +25,7 @@ export default function StatusesTable(props) {
                     <tr>
                         <th scope="col">Rank</th>
                         <th scope="col">Status Text</th>
-                        <th scope="col">Retweeter Count</th>
+                        <th scope="col">Bot Count</th>
                     </tr>
                 </thead>
                 <tbody>

src/_App/BotAnalysis/Classification/NOTES.md

@@ -1,5 +1,21 @@
 
 
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+## Daily Bot Probabilities Histogram
+
 ```sql
 
 /*
@@ -27,3 +43,11 @@ So in BigQuery only the users with bot probabilities above 0.5 have been saved.
 Need to dig into the google cloud storage bucket for the full CSV file.
 
 See [API Prep](https://github.com/s2t2/tweet-analyzer-py/pull/65/files) and run the script to generate the JSON file. Then copy the JSON file into this dir.
+
+
+
+
+## Bots Most Active
+
+```sql
+```

src/_App/BotAnalysis/Classification/Section.js

@@ -28,14 +28,13 @@ export default function BotClassificationSection() {
         <Container fluid>
             <Card>
                 <Card.Body>
-                    <Card.Title><h3>Bot Classification</h3></Card.Title>
+                    <Card.Title><h3>Bot Detection</h3></Card.Title>
 
-                    <h4 id="bot-behaviors">Bot Behaviors</h4>
+                    <h4 id="retweet-analysis">Retweet Analysis</h4>
                     <Card.Text>
                         The <a href="/about">previous bot detection research</a> suggests bots exhibit the primary behavior of retweeting humans at high frequencies.
                     </Card.Text>
 
-                    <h4 id="retweet-analysis">Retweet Analysis</h4>
                     <Card.Text>
                         We examined the retweets for each day in our <a href="/collection-timeline">collection period</a> to identify which users retweeted with sufficient frequency to differentiate them from humans. Based on these retweet frequencies, our bot classification model assigned each retweeter a "daily bot probability score" from <code>0</code> (not bot) to <code>1</code> (bot). An example distribution of daily bot probability scores is below.
                     </Card.Text>

src/_App/BotAnalysis/Impact/Section.js

@@ -18,22 +18,33 @@ export default function BotImpact() {
                     <Card.Title><h3>Bot Impact</h3></Card.Title>
 
                     <h4>Opinion Shift</h4>
+
+                    <DailyShift/>
+
+                    {/*
                     <Card.Text>
                         For each day in our <a href="/collection-timeline">primary tweet collection period</a>,
                         {" "} we used our <a href="/opinion-models">Impeachment opinion model</a>
                         {" "} to calculate the average opinion scores for all users, with vs. without <a href="/bot-classification">bots</a>,
                         {" "} to assess the impact the bots were having on the conversation.
+                    </Card.Text>
+                    */}
 
+                    <Card.Text>
+                        The chart above shows the daily bot-induced shift in opinions about the Impeachment of President Donald Trump.
+                        {" "}The shift in mean opinion is affected by bot reach, <a href="/bot-activity">bot activity</a> levels, and <a href="/bot-beneficiaries">bot opinions</a>.
+                        {" "}Baseline opinions are measured using our <a href="/opinion-models">Impeachment opinion model</a>, which is based on a BERT transformer sentiment classifier.
                     </Card.Text>
+
+
                     <Card.Text>
                         {" "} We observe the average daily bot-induced opinion shift is 1.4% towards left-leaning opinions.
-                        {" "} The chart below shows the shift in opinion scores by day.
                         {" "} The highest daily shift towards right-leaning bot opinions was 2.8% on 12/28, and the highest daily shift towards left-leaning bot opinions was 6.3% on 1/8.
                     </Card.Text>
+
                     {/*
                     <img src={dailyShift} alt="a bar graph of daily opinion shift by bot community" style={{marginTop:20, marginBottom:20}} className="img-fluid"/>
                     */}
-                    <DailyShift/>
 
                 </Card.Body>
             </Card>

src/_App/BotAnalysis/Page.js

@@ -15,9 +15,9 @@ export default function BotAnalysisPage() {
             <SectionIndex/>
 
             <ImpactSection/>
-            <ActivitySection/>
             <BeneficiariesSection/>
             <LanguageSection/>
+            <ActivitySection/>
             <CommunitiesSection/>
             <ClassificationSection/>
         </span>

src/_App/Home.js

@@ -6,10 +6,10 @@ import React from 'react'
 //import Card from 'react-bootstrap/Card'
 
 //import HomePage from './OpinionAnalysis/Page'
-import HomePage from './BotAnalysis/Page'
+//import HomePage from './BotAnalysis/Impact/Page'
 
 //import AboutCard from './About/Card'
-//import Section from './BotAnalysis/Impact/Section'
+import HomePage from './BotAnalysis/Impact/Section'
 
 export default function Home() {
 

src/_App/sidebar.js

@@ -20,6 +20,19 @@ import UserOpinions from './OpinionAnalysis/User/Page'
 import TopUserOpinions from './OpinionAnalysis/TopUsers/Page'
 
 var sidebar = [
+    {
+        "key": "bot-analysis",
+        "title": "Bot Analysis",
+        "component": BotAnalysis,
+        "sections": [
+            {"key": "bot-impact",           "title": "Bot Impact",          "component": BotImpact},
+            {"key": "bot-beneficiaries",    "title": "Bot Beneficiaries",   "component": BotBeneficiaries},
+            {"key": "bot-language",         "title": "Bot Language",        "component": BotLanguage},
+            {"key": "bot-activity",         "title": "Bot Activity",        "component": BotActivity},
+            {"key": "bot-communities",      "title": "Bot Communities",    "component": BotCommunities},
+            {"key": "bot-detection",        "title": "Bot Detection",      "component": BotClassification}
+        ]
+    },
     {
         "key": "opinion-analysis",
         "title": "Opinion Analysis",
@@ -30,19 +43,6 @@ var sidebar = [
             {"key": "opinion-models",       "title": "Opinion Models",      "component": OpinionModels},
         ]
     },
-    {
-        "key": "bot-analysis",
-        "title": "Bot Analysis",
-        "component": BotAnalysis,
-        "sections": [
-            {"key": "bot-impact",           "title": "Bot Impact",          "component": BotImpact},
-            {"key": "bot-activity",         "title": "Bot Activity",        "component": BotActivity},
-            {"key": "bot-beneficiaries",    "title": "Bot Beneficiaries",  "component": BotBeneficiaries},
-            {"key": "bot-language",         "title": "Bot Language",        "component": BotLanguage},
-            {"key": "bot-communities",       "title": "Bot Communities",    "component": BotCommunities},
-            {"key": "bot-classification",    "title": "Bot Classification",  "component": BotClassification}
-        ]
-    },
     {
         "key": "tweet-collection",
         "title": "Tweet Collection",