Sentiment Data Analysis
Budget: $30 – $250 USD
I am looking for a freelancer who can help me with sentiment data analysis specifically for social media posts on Instagram.
Background - I am studying the impact of influencer posts on the overall brand image in context of paid influencer marketing campaigns. This is being analyzed through data mining based on comments by followers on influencer posts (only for branded content), and deriving the sentiment at a campaign level.
Data availability: I will provide social media campaign analytics data for a specific consumer category (Personal Care and Beauty) at brand level (11 brands), along with specific posts created by influencers employed by brands for a specific period. The campaign data (about 20k unique posts) is for a period of 1 year, and each campaign can be identified based on unique hashtag, as well as date and time stamp. In addition, the analytics data captures URLs for specific influencer posts, along with the social media platform (Instagram and YouTube) on which these were posted.
Requirement: Each individual post needs to be reviewed, all comments in response to the post should be scraped, and analyzed for sentiment. This should include English, Hinglish, and interpretation of emojis to arrive at overall sentiment.
In addition, the following analysis should be done –
1. Since every campaign deploys multiple influencers, analysis of the comments to measure sentiment must be done at post level (for finite and unique posts), influencer level (for finite and unique posts by specific influencers), and at campaign level (by hashtag and time stamp), and specific to each social media platform (Instagram and YouTube)
2. It should also be explored if Likes/Dislikes and Reshares have a bearing on the overall sentiment of the post.
3. Based on the nature/format of content – that is image, video, gif, plain text, (and combinations of image + text; video + text; gif + text) sentiments should be computed – The idea is to understand if format of the content has any impact on the post sentiment.
Kindly write "sentiment analysis" at the top of your proposal.
This is going to be long term work for ongoing various research projects.
Background - I am studying the impact of influencer posts on the overall brand image in context of paid influencer marketing campaigns. This is being analyzed through data mining based on comments by followers on influencer posts (only for branded content), and deriving the sentiment at a campaign level.
Data availability: I will provide social media campaign analytics data for a specific consumer category (Personal Care and Beauty) at brand level (11 brands), along with specific posts created by influencers employed by brands for a specific period. The campaign data (about 20k unique posts) is for a period of 1 year, and each campaign can be identified based on unique hashtag, as well as date and time stamp. In addition, the analytics data captures URLs for specific influencer posts, along with the social media platform (Instagram and YouTube) on which these were posted.
Requirement: Each individual post needs to be reviewed, all comments in response to the post should be scraped, and analyzed for sentiment. This should include English, Hinglish, and interpretation of emojis to arrive at overall sentiment.
In addition, the following analysis should be done –
1. Since every campaign deploys multiple influencers, analysis of the comments to measure sentiment must be done at post level (for finite and unique posts), influencer level (for finite and unique posts by specific influencers), and at campaign level (by hashtag and time stamp), and specific to each social media platform (Instagram and YouTube)
2. It should also be explored if Likes/Dislikes and Reshares have a bearing on the overall sentiment of the post.
3. Based on the nature/format of content – that is image, video, gif, plain text, (and combinations of image + text; video + text; gif + text) sentiments should be computed – The idea is to understand if format of the content has any impact on the post sentiment.
Kindly write "sentiment analysis" at the top of your proposal.
This is going to be long term work for ongoing various research projects.