Customer Analytics with Social Media

Added on -2020-02-19

| 24 pages| 3062 words| 222 views

Trusted by 2+ million users,
1000+ happy students everyday

Showing pages 1 to 6 of 24 pages

Customer Analytics with Social Media (Assignment 1 - Social Media Analysis for Understanding CustomerPreferences and Sentiments)By<Student Name>(18833953)
Table of Contents1.Introduction12.Case Study: 122.1Properties of articles2Number of shares22.2Keyword analysis using SAS Text Miner63.Case Study: 210AppendixA
List of FiguresFigure 1 Title length (Overall): Number of shares..........................................................................3Figure 2 Content length (Overall) : Number of shares....................................................................4Figure 3 Published on the weekend (Overall) : Number of shares..................................................4Figure 4 Top 10 shares : Average title lengths................................................................................5Figure 5 Top 10 shares : Average content lengths..........................................................................5Figure 6 Top 10 shares : Published in weekdays:............................................................................6Figure 7 Topics of whole data vs Share.........................................................................................10Figure 9 qplot of sentiments..........................................................................................................11
List of TablesTable 1 “Lifestyle” channel : Top 10 topics6Table 2c“Entertainment” channel : Top 10 topics7Table 3 “Business” channel: Top 10 topics7Table 4 “Social Media” channel : Top 10 topics8Table 5 “Technology” channel : Top 10 topics8Table 6 “World” channel : Top 10 topics9Table 7 “Whole” data : Top 10 topics9
1.IntroductionText mining can be mentioned as a special type of practice and strategy which employs the rules of datamining to text. It is an automated process which helps us to identify and disclose previously undiscovereddesigns of text data. Sentiment analysis, assists us to obtain the attitudes, feelings, and views ofindividuals and groups from textual data and contents. Sentiment analysis is usually applied to sentencesas well as short messages. Sentiment analysis can also be implemented to the entire passage to assess theimportance of a view or outlook1. Combination of text mining and sentiment analysis can providesuperior power to describe and predict any textual content of social media behavior2, significant amountof descriptive and predictive power we are provided with a significant amount of descriptive andpredictive power3. In this study, our main purposes are to find the impact of articles depending uponsharing and understand movie reviewers’ sentiment in social networking site and comparison of emotionsusing different text mining methods. In case study 1, our aim is to discover the connection amongdifferent keywords, number of shares, and different expressions, which would assist to formulate strategyto improve advertise campaigning and improved communication with customers. In case study 2, our aimis to understand sentiments of consumers’ review, which would assist to find insights to enhance thecampaign tracking.1 Khan, F. H., Bashir, S., & Qamar, U. (2014). TOM: Twitter opinion mining framework using hybridclassification scheme. Decision Support Systems, 57, 245-257.2 Khan, F. H., Bashir, S., & Qamar, U. (2014). TOM: Twitter opinion mining framework using hybridclassification scheme. Decision Support Systems, 57, 245-257.3 Khan, F. H., Bashir, S., & Qamar, U. (2014). TOM: Twitter opinion mining framework using hybridclassification scheme. Decision Support Systems, 57, 245-257.1
2.Case Study: 11.1Properties of articlesThere are 2100 observations for life style, 7059 for entertainment, 6259 for business, 2325 for socialmedia, 7345 for technology, 8425 for world, and 6147 datasets without association with any of thesechannels. Number of sharesOverall highest number of shares i.e. 843300, were for “Leaked: More Low-Cost iPhone Photos”, (this isnot falling under above mentioned six channels). Highest number of shares under lifestyle channel have been observed for “Obama to Discuss NSAReform with Lawmakers” i.e. 208300. Under entertainment channel, maximum number of shares have been observed for “Sprint's New PlansGuarantee Unlimited Data for Life” i.e. 210300. Under business channel, maximum number of shares have been observed for “Dove Experiment Aims toChange the Way You See Yourself” i.e. 690400. Under social media channel, maximum number of shares have been observed for “World's First Sprout-Powered Battery Just Lit Up a Christmas Tree [VIDEO]i.e. 122800.Under technology channel, maximum number of shares have been observed for “Startup stories fromearly hiresi.e. 663600. Under world channel, maximum number of shares have been observed for “U.S. Will Now Monitor AllTravelers From Ebola Zone for 21 Daysi.e. 284700. 2

Found this document preview useful?

You are reading a preview
Upload your documents to download
or
Become a Desklib member to get accesss

Premium

$45

Q&A Library Access

Chat support

12

Document Unlocks

4

Answer Unlocks

Students who viewed this