Chapter 4: SMS case study - invalid input

[jingo99]

Dear @dataspelunking,

I am a Dutch student currently working through your book for a project. I'm an absolute beginner at machine learning and your book is very helpful in teaching me. At chapter 4 page 117, I'm instructed to create 2 wordclouds based on the raw data, so I can inspect the difference in "spam" and "ham". However, when creating the wordclouds, my RStudio fails and tells me the following.

s_raw <- read.csv("sms_spam.csv", stringsAsFactors = FALSE)
wordcloud(spam$text, max.words = 40, scale = c(3, 0.5))

transformation drops documentstransformation drops documentsError in strwidth(words[i], cex = size[i], ...) : invalid input 'â£1000' in 'utf8towcs'

I'm a student developer, so I presumed it might have been the (british) pound-sign failing the code. I retrieved your updated csv-file from github, but to no avail. I have continued working, but if possible, i'd like to understand why this code fails.

I'm not able to really ask anyone for help, so I thought I'd try my luck here.

Would you be so kind as to explain what i'm missing here?

With kind regards,

Jingo