Topic Modeling

Chris Bail
Duke University

website: https://www.chrisbail.net
Twitter: https://www.twitter.com/chris_bail
github: https://github.com/cbail

What is Topic Modeling?

Latent Dirichlet Allocation

Example: LDA of Scientific Abstracts

Running Your First Topic Model

library(topicmodels)
library(tm)

data("AssociatedPress")

inspect(AssociatedPress[1:5, 1:5])

<<DocumentTermMatrix (documents: 5, terms: 5)>>
Non-/sparse entries: 0/25
Sparsity           : 100%
Maximal term length: 10
Weighting          : term frequency (tf)
Sample             :
      Terms
Docs   aaron abandon abandoned abandoning abbott
  [1,]     0       0         0          0      0
  [2,]     0       0         0          0      0
  [3,]     0       0         0          0      0
  [4,]     0       0         0          0      0
  [5,]     0       0         0          0      0

Running Your First Topic Model

AP_topic_model<-LDA(AssociatedPress, 
                    k=10, 
                    control = list(seed = 321))

Running Your First Topic Model

library(tidytext)
library(dplyr)

AP_topics <- tidy(AP_topic_model, matrix = "beta")

ap_top_terms <- 
  AP_topics %>%
    group_by(topic) %>%
      top_n(10, beta) %>%
        ungroup() %>%
          arrange(topic, -beta)

Plot

library(ggplot2)

ap_top_terms %>%
  mutate(term = reorder(term, beta)) %>%
    mutate(topic = paste("Topic #", topic)) %>%
    ggplot(aes(term, beta, fill = factor(topic))) +
      geom_col(show.legend = FALSE) +
        facet_wrap(~ topic, scales = "free") +
          theme_minimal()+
          theme(plot.title = 
            element_text(hjust = 0.5, size=18))+
          labs(
            title = "Topic Model of AP News Articles",
            caption = "Top Terms by Topic (betas)"
          )+
          ylab("")+
          xlab("")+
          coord_flip()

Plot

Reading Tea Leaves

Structural Topic Modeling