Difference between revisions of "ANLY482 AY2017-18T2 Group30 Facebook Video"

From Analytics Practicum
Jump to navigation Jump to search
(Created page with "<!--Team Logo--> center|300px| <!--End of Team Logo--> <br/> <!--Main Navigation--> <center> {|style="background-color:#5A6B96; color:#5A6B96; width=...")
 
 
(9 intermediate revisions by the same user not shown)
Line 9: Line 9:
 
{|style="background-color:#5A6B96; color:#5A6B96; width="100%" cellspacing="0" cellpadding="10" border="0" |
 
{|style="background-color:#5A6B96; color:#5A6B96; width="100%" cellspacing="0" cellpadding="10" border="0" |
  
−
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="14%" | [[ANLY482_AY2017-18_T2_Group_30|<font color="#FFFFFF" size=3><b>HOME</b></font>]]
+
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="12.5%" | [[ANLY482_AY2017-18_T2_Group_30|<font color="#FFFFFF" size=3><b>HOME</b></font>]]
  
−
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="14%" | [[ANLY482_AY2017-18T2_Group30 About Us |<font color="#FFFFFF" size=3><b>ABOUT US</b></font>]]
+
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="12.5%" | [[ANLY482_AY2017-18T2_Group30 About Us |<font color="#FFFFFF" size=3><b>ABOUT US</b></font>]]
  
−
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="14%" | [[ANLY482_AY2017-18T2_Group30 Project Overview |<font color="#FFFFFF" size=3><b>PROJECT OVERVIEW </b></font>]]
+
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="12.5%" | [[ANLY482_AY2017-18T2_Group30 Project Overview |<font color="#FFFFFF" size=3><b>PROJECT OVERVIEW </b></font>]]
  
−
|style="font-size:88%; border-left:1px solid #347cc4; border-right:1px solid #347cc4; text-align:center; border-bottom:1px solid #347cc4; border-top:1px solid #347cc4;" width="14%" |[[ANLY482_AY2017-18T2_Group30 Data Analysis |<font color="#347cc4" size=3><b>PROJECT FINDINGS</b></font>]]
+
|style="font-size:88%; border-left:1px solid #347cc4; border-right:1px solid #347cc4; text-align:center; border-bottom:1px solid #347cc4; border-top:1px solid #347cc4;" width="12.5%" |[[ANLY482_AY2017-18T2_Group30 Data Analysis |<font color="#347cc4" size=3><b>EDA</b></font>]]
  
−
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="14%" | [[ANLY482_AY2017-18T2_Group30 Project Management |<font color="#FFFFFF" size=3><b>PROJECT MANAGEMENT</b></font>]]
+
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="12.5%" | [[ANLY482_AY2017-18T2_Group30 Business Objectives |<font color="#FFFFFF" size=3><b>BUSINESS OBJECTIVES </b></font>]]
  
−
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="15%" |[[ANLY482_AY2017-18T2_Group30 Documentation | <font color="#FFFFFF" size=3><b>DOCUMENTATION</b></font>]]
+
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="12.5%" | [[ANLY482_AY2017-18T2_Group30 Project Management |<font color="#FFFFFF" size=3><b>PROJECT MANAGEMENT</b></font>]]
  
−
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="15%" |[[ANLY482_AY2017-18_Term_2 | <font color="#FFFFFF" size=3><b>MAIN PAGE</b></font>]]
+
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="12.5%" |[[ANLY482_AY2017-18T2_Group30 Documentation | <font color="#FFFFFF" size=3><b>DOCUMENTATION</b></font>]]
 +
 
 +
|style="font-size:88%; border-left:1px solid #ffffff; border-right:1px solid #ffffff; text-align:center; background-color:#347cc4; " width="12.5%" |[[ANLY482_AY2017-18_Term_2 | <font color="#FFFFFF" size=3><b>MAIN PAGE</b></font>]]
 
|}  
 
|}  
 
</center>
 
</center>
Line 47: Line 49:
 
<!---------------END of sub menu ---------------------->
 
<!---------------END of sub menu ---------------------->
 
<br>
 
<br>
−
<div align="center">
 
−
<div style=" width: 85%; background: #E6EDFA; padding: 12px; font-family: Arimo; font-size: 18px; font-weight: bold; line-height: 1em; text-indent: 15px; border-left: #8c8d94 solid 32px;"><font color="#5A6B96">Data Source</font></div>
 
−
<div style="width:90%;">
 
−
<font style="text-align: left">
 
−
<p>
 
−
<b>Facebook</b><br>
 
−
For data files from <i>Facebook Insights Data Export (Post Level)</i>, the sponsor provided exported data from different periods of the year, with different metric tabs in Excel format.  The tabs included are:
 
−
# Key Metrics
 
−
# Lifetime: Number of unique people who have created a story about your Page post by interacting with it (unique users)
 
−
# Lifetime: Number of people who have clicked anywhere in your post, by type (unique users)
 
−
# Lifetime: Number of people who have given negative feedback on your post, by type (unique users)
 
−
<br></p>
 
  
−
<font style="text-align: left">
+
==<div style=" width: 96.5%; background: #E6EDFA; padding: 12px; font-family: Arimo; font-size: 18px; font-weight: bold; line-height: 1em; text-indent: 15px; border-left: #8c8d94 solid 32px;"><font color="#5A6B96">Data Source</font></div>==
−
<p>
+
 
−
For data files from <i>Facebook Insights Data Export (Video Post)</i>, the sponsor provided exported data from different periods of the year, with different metric tabs in Excel format.  The tabs included are:  
+
<div style="width:96.5%;">
 +
<font>
 +
For data files from Facebook Insights Data Export (Video Post), the sponsor provided exported data from different periods of the year, with different metric tabs in Excel format.  The tabs included are:  
 
# Lifetime Post Total Impression/Reach/Views
 
# Lifetime Post Total Impression/Reach/Views
 
# Geographic Views
 
# Geographic Views
 
# Demographic Views
 
# Demographic Views
 
# Lifetime Post Toal Views by (page_owned / Shared)
 
# Lifetime Post Toal Views by (page_owned / Shared)
−
</p></font>
 
−
 
−
<div style="width:90%;">
 
−
<font style="text-align: left">
 
−
<b>YouTube</b><br/>
 
−
For data files from <i>YouTube(Watch Time)</i>, the sponsor provided exported data for Watch Time, with different metric tabs in Excel format. The tabs included are:
 
−
# Video
 
−
#Geography
 
−
#Date
 
−
#Subscription Status
 
−
#Youtube Product
 
−
#Device Type
 
−
#Subtitles and CC
 
−
#Video Information Language
 
−
<br>
 
−
For data files from <i>YouTube(Demographics)</i>, the sponsor provided exported data for watch time for different Demographic, with different metric tabs in Excel format. The tabs included are:
 
−
# Viewer Age
 
−
# Viewer Gender
 
−
<br>
 
−
For data files from <i>YouTube(Traffic Sources)</i>, the sponsor provided exported data for watch time from different traffic source type
 
 
</font>
 
</font>
 
</div>
 
</div>
  
−
<br/><b>Instagram</b><br/>
 
−
To retrieve data from the company's instagram, we made use of a web-scraping script from [https://github.com/timgrossmann/instagram-profilecrawl Github]. We made modifications to the script to include timestamp as well as caption, the data includes:
 
−
* Caption
 
−
* Timestamp
 
−
* Img URL
 
−
* Tags
 
−
* No. of Likes
 
−
* No. of Comments
 
  
−
<br/><b>Blog</b><br/>
+
==<div style=" width: 96.5%; background: #E6EDFA; padding: 12px; font-family: Arimo; font-size: 18px; font-weight: bold; line-height: 1em; text-indent: 15px; border-left: #8c8d94 solid 32px;"><font color="#5A6B96">Data Preparation</font></div>==
−
To retrieve data from the company's posts, we used [https://scrapy.org/ Scrapy], a fast and powerful open-sourced web-scraper to extract data from the blog. We collected data from the beginning of the first blog post, with the following information:
 
−
* Timestamp
 
−
* Author(s)
 
−
* Headline
 
−
* Category
 
−
* URL
 
−
* Tags
 
  
−
</font></div>
+
<div style="width:96.5%;">
−
<br/>
+
<div align="left">
 +
[[File:Fbvideo clean1.PNG|FB Clean 1|700px]]
 +
[[File:Fbvideo clean2.PNG|FB Clean 2|700px]]
 +
[[File:Fbvideo clean3.PNG|FB Clean 3|700px]]
 +
</div>
 +
</div>
 +
 
 +
==<div style=" width: 96.5%; background: #E6EDFA; padding: 12px; font-family: Arimo; font-size: 18px; font-weight: bold; line-height: 1em; text-indent: 15px; border-left: #8c8d94 solid 32px;"><font color="#5A6B96">Exploratory Data Analysis</font></div>==
  
 
<div align="center">
 
<div align="center">
−
<div style=" width: 85%; background: #E6EDFA; padding: 12px; font-family: Arimo; font-size: 18px; font-weight: bold; line-height: 1em; text-indent: 15px; border-left: #8c8d94 solid 32px;"><font color="#5A6B96">Data Preparation</font></div>
+
[[File:Dailyviews.png|Video Views for first 30 days|1000px]]
−
<div style="width:90%;">
+
</div>
−
<font style="text-align: left">
 
−
<p>
 
−
To help us have an overview of the data throughout the year, we consolidated the various tabs, whilst concatenating the various periods of data for the same columns, into one combined file. This was carried out using the software, IBM JMP Pro, in the following steps:
 
−
* With Post ID, Permalink (permanent link of the campaign content), Post Message, Type, Countries and Posted columns as key identifiers among the different tabs for the excel files, we appended desired columns from the other tabs to the end of the Key Metrics. They included the Share, Like, Comment columns from Tab 2; Other Clicks, Link Clicks, Photo View, Video Play columns from Tab 3; Hide_Clicks , Hide_all_clicks, Unlike_page_clicks, report_spam_clicks columns from Tab 4. <br>This was conducted using the <i>Tables > Join </i>function, with “Matching Specification” as the key identifiers and “Output Columns” of the appended desired columns.
 
  
−
* Next, for each period of data files (appended with new columns) from multiple tabs, we concatenate the data across different time periods to have a full year collection of data.<br>This was conducted using the <i>Tables > Concatenate </i> function, while adding multiple data tables into “Data Tables to be Concatenated”.  
+
<div style="width:96.5%;">
 +
<font>
 +
The diagram above demonstrates the number of views for a video from Day 1 to Day 30. It would be interesting to note which are the videos that have more views than the upper whisker for each day. These videos would be termed as high-performance videos. Most of the views are generated within the first two days and view count would decrease significantly from day 3 onwards.  
  
−
* Finally, we check for missing data in the different columns. For example, under the column Type, we have five different types, namely: Link, Photo, Shared Video, Status and Video. However, in the instances of missing data, we will cross check with the permalink of the campaign post, and check the Type of medium was posted and fill it in accordingly.  
+
In subsequent analysis, we aim to categorize videos into different topics so that we will have an understanding of which topics that are popular amongst the Singapore population. The current bar chart just displays the ranking of videos in a descending order with its respective video tags. The video viewership demographics shows the type of audience that has viewed the video.
−
</p>
+
</font>
−
</font></div>
+
</div>
−
<br/>
 
  
 
<div align="center">
 
<div align="center">
−
<div style=" width: 85%; background: #E6EDFA; padding: 12px; font-family: Arimo; font-size: 18px; font-weight: bold; line-height: 1em; text-indent: 15px; border-left: #8c8d94 solid 32px;"><font color="#5A6B96">Data Cleaning</font></div>
+
[[File:Weekday views2.png|Weekday views|1000px]]
−
<div style="width:90%;">
+
</div>
−
 
 
−
<div align="left">
 
−
<b>Instagram Data</b><br/>
 
−
After scraping the data, we realised that the data needed cleaning. The indexes of the column values were off as seen here:
 
−
(image)
 
−
We also concatenated the "tags" into a single column.
 
−
</div></div>
 
−
 
 
−
 
 
−
<div style=" width: 85%; background: #E6EDFA; padding: 12px; font-family: Arimo; font-size: 18px; font-weight: bold; line-height: 1em; text-indent: 15px; border-left: #8c8d94 solid 32px;"><font color="#5A6B96">Exploratory Data Analysis</font></div>
 
−
<br/>
 
−
 
 
−
 
 
−
<div style=" width: 85%; background: #E6EDFA; padding: 12px; font-family: Arimo; font-size: 18px; font-weight: bold; line-height: 1em; text-indent: 15px; border-left: #8c8d94 solid 32px;"><font color="#5A6B96">Final Application: Learning Dashboard</font></div>
 
−
<br/>
 
  
 +
<div style="width:96.5%;">
 +
<font>
 +
Graph on the left shows the number of videos that were posted across the week. We can observe that more videos were posted across the weekday from Monday to Thursday, followed by a decrease in videos posted from Friday to Sunday.
 +
(Hypothesis 1)Hence we would assume that viewership would have the same trend across the same period of time. However, this doesn't seem to be the case.
 +
A graph on the right shows the average number of views per weekday. Views are generally higher on Monday and viewership would decline across the week, with an exception on Friday, where viewership has a slight spike followed by a decrease towards Sunday. This is a striking contrast against graph on the left and proved our hypothesis false. Our team believes that viewership count is perpetuated by other circumstances other than the amount of videos posted.
 +
</font>
 
</div>
 
</div>

Latest revision as of 13:26, 10 April 2018

APex Logo.PNG


HOME ABOUT US PROJECT OVERVIEW EDA BUSINESS OBJECTIVES PROJECT MANAGEMENT DOCUMENTATION MAIN PAGE
Facebook Post Facebook Video Youtube Instagram Blog Post


Data Source

For data files from Facebook Insights Data Export (Video Post), the sponsor provided exported data from different periods of the year, with different metric tabs in Excel format. The tabs included are:

  1. Lifetime Post Total Impression/Reach/Views
  2. Geographic Views
  3. Demographic Views
  4. Lifetime Post Toal Views by (page_owned / Shared)


Data Preparation

FB Clean 1 FB Clean 2 FB Clean 3

Exploratory Data Analysis

Video Views for first 30 days

The diagram above demonstrates the number of views for a video from Day 1 to Day 30. It would be interesting to note which are the videos that have more views than the upper whisker for each day. These videos would be termed as high-performance videos. Most of the views are generated within the first two days and view count would decrease significantly from day 3 onwards.

In subsequent analysis, we aim to categorize videos into different topics so that we will have an understanding of which topics that are popular amongst the Singapore population. The current bar chart just displays the ranking of videos in a descending order with its respective video tags. The video viewership demographics shows the type of audience that has viewed the video.

Weekday views

Graph on the left shows the number of videos that were posted across the week. We can observe that more videos were posted across the weekday from Monday to Thursday, followed by a decrease in videos posted from Friday to Sunday. (Hypothesis 1)Hence we would assume that viewership would have the same trend across the same period of time. However, this doesn't seem to be the case. A graph on the right shows the average number of views per weekday. Views are generally higher on Monday and viewership would decline across the week, with an exception on Friday, where viewership has a slight spike followed by a decrease towards Sunday. This is a striking contrast against graph on the left and proved our hypothesis false. Our team believes that viewership count is perpetuated by other circumstances other than the amount of videos posted.

Retrieved from ‘https://wiki.smu.edu.sg/ANLY482/index.php?title=ANLY482_AY2017-18T2_Group30_Facebook_Video&oldid=22841’