BackgroundCoca-Cola launched four new Diet Coke flavors in January 2018: Ginger Lime, Feisty Cherry,Zesty Blood Orange, and Twisted Mango. The new flavors were accompanied by modernizedpackaging and a new ad campaign. Due to stagnation in sales with Millennials, Diet Cokesought to use this campaign as a way to attract Millennial attention.The account team has asked the analytics team to perform a series of analysis to determinethe effectiveness of the campaign. This includes but is not limited to: identifying areas ofimprovement, determining whether the target audience was reached, and evaluating how welltheir set of product influencers performed.As a data analyst, you are asked to use python(2or3) to perform the following dataanalysis to support business team. Your final deliverables will include: spreadsheet(csv)files reflecting all the work you would have done and notebooks with your code togenerate the csv files.AnalysisUse Diet Coke Raw Data.csv for the following analysis.Section 1 Metrics1. We define “social engagement” as the sum number of “likes”, “comments” and“retweet”. Identify the volume of mentions and total engagement for each month foreach platform and output the result in csv format.2. Only keep the mentions from Twitter, Instagram, Facebook, News, Blogs and Forums,and group all the related mentions into 3 categories: 1) Social, 2) News, 3) Blog &Forum. Then try to identify which category has the highest sentiment and what’s thetotal number of positive mentions of that group? Output the result in csv format.3. Find the top 10 authors with the highest followers for each social platform. Businessanalyst also want a Url link to authors profile or mention page, help them on that aswell. Output the result in csv format.4. List all the news websites. (i.e. NYTimes)Section 2 Topic Analysis1. Write functions to find the official posts for each brand. Add a new column called“Official” and tag the brand name for each official mention. All the official handles canbe found in Brand Official Handle Keywords.csv.2. Following step 1, tag all the mentions left as “UGC”(User Generated Content). Writefunctions to identify topic for each UGC mention in only “Social” category by usingEmotions Keywords.csv. (The first row of the file is the topic name, from the second rowand below are the keywords for each topic) Tag topic name for each mention beingmatched.○ (Hint) It is possible that some mentions can be matched by multiple topics. Soyou need to keep all matched topics.○ (Hint) Try to use regular expression. Explain the advantages and disadvantagesof regular expression.○ (Hint) Make sure to take care of case sensitivity.3. Find the number of mentions, total engagement and positive sentiment percentage foreach topic.Section 3 Longform AnalysisIn this section, you will be asked to complete the topic analysis and sentiment analysis onlongform posts. We define all the mentions from News, Blogs, and Forums to be longformposts. The difficulty of dealing with longform texts is one text can talk about differenttopics in different segments. To illustrate, take a look at the example below. This piece oftext talks topic “Surprise” twice. The first appearance is captured by “thrill” and second by“amazing”. Each highlighted part correspond to one segment.In order to calculate the sentiment as accurate as possible, we are going to do thefollowing:1. Segment the post into individual sentences.2. Match each sentences against keyword.3. The matched sentence with 2 sentences before and 2 after combined will be markedas “key content”.Look back at the example above, this is how the example is generated.Similar as section 2, in this section, we want you to write functions to complete the followingsteps.1. Classify topics and attach key contents1.1. Classify topics for all mentions from News and Blog & Forumcategories using Emotions Keywords.csv1.2. Attach all key contents for each topic in each long form article.(For instance, create a new column called “key content {topic}”for each topic in the article. Suppose that there are 5 topics inone article, you can either append 5 “key content {topic}”column for this article, or create additional 5 rows, each rowhave one different topic and the associated key contents. Eachtopic should be treated independently.)2. Find the sentiment score for every emotioned topic in each longformarticles.Hint: you may use this https://github.com/youhealthy/vaderSentiment3. Please included the number of topics matched for each longform article.Hint: It is possible that some longform articles can be matched by multiple emotion topics.Section 4 DashboardIn this section, you are required to complete the data visualization for the following charts in aDashboard.1. Make a line graph to indicate the weekly volume of mentions change in the data set.2. Use Bar chart to list top 10 websites by volume of mentions for News categoryand Blog & Forum categories respectively.3. Create a bubble chart to show the engagement weekly change in the Social category.The x-axis would be the week, the y-axis would be the percentage of positive mentionsof the week, and the bubble size would be engagement.本团队核心人员组成主要包括硅谷工程师、BAT一线工程师，精通德英语！我们主要业务范围是代做编程大作业、课程设计等等。我们的方向领域：window编程 数值算法 AI人工智能 金融统计 计量分析 大数据 网络编程 WEB编程 通讯编程 游戏编程多媒体linux 外挂编程 程序API图像处理 嵌入式/单片机 数据库编程 控制台 进程与线程 网络安全 汇编语言 硬件编程 软件设计 工程标准规等。其中代写编程、代写程序、代写留学生程序作业语言或工具包括但不限于以下范围:C/C++/C#代写Java代写IT代写Python代写辅导编程作业Matlab代写Haskell代写Processing代写Linux环境搭建Rust代写Data Structure Assginment 数据结构代写MIPS代写Machine Learning 作业 代写Oracle/SQL/PostgreSQL/Pig 数据库代写/代做/辅导Web开发、网站开发、网站作业ASP.NET网站开发Finance Insurace Statistics统计、回归、迭代Prolog代写Computer Computational method代做因为专业，所以值得信赖。如有需要，请加QQ：99515681 或邮箱：99515681@qq.com 微信：codehelp
网友评论