Site icon R-bloggers

Better Sentiment Analysis with sentiment.ai

[This article was first published on Happy Little Scripts, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

 

< content="text/html;charset=UTF-8" http-equiv="Content-type">
< !-- [START BADGES] -->< !-- Please keep comment here to allow auto update -->

< !-- [END BADGES] -->

sentiment.ai

More powerful sentiment analysis using deep learning models made simple..

See github.io page herehttps://benwiseman.github.io/sentiment.ai/

Overview

Korn Ferry Institute‘s AITMI team made sentiment.ai for researchers and tinkerers who want a straight-forward way to use powerful, open source deep learning models to improve their sentiment analyses. Our approach is relatively simple and out performs the current best offerings on CRAN and even Microsoft’s Azure Cognitive Services. Given that we felt the current norm for sentiment analysis isn’t quite good enough, we decided to open-source our simplified interface to turn Universal Sentence Encoder embedding vectors into sentiment scores.

We’ve wrapped a lot of the underlying hassle up to make the process as simple as possible. In addition to just being cool, this approach solves several problems with traditional sentiment analysis, namely:

  1. More robust, can handle spelling mitsakes and mixed case, and can be applied to dieciséis (16) languages!

  2. Doesn’t need a ridged lexicon, rather it matches to an embedding vector (reduces language to a vector of numbers that capture the information, kind of like a PCA). This means you can get scores for words that are not in the lexicon but are similar to existing words!

  3. Choose the context for what negative and positive mean using the sentiment_match() function. For example, you could set positive to mean "high quality" and negative to mean "low quality" when looking at product reviews.

  4. Power Because it draws from language embedding models trained on billions of texts, news articles, and wikipedia entries, it is able to detect things such as “I learned so much on my trip to Hiroshima museum last year!” is associated with something positive and that “What happeded to the people of Hiroshima in 1945” is associated with something negative.

  5. The power is yours We’ve designed sentiment.ai such that the community can contribute sentiment models via github. This way, it’s easier for the community to work together to make sentiment analysis more reliable! Currently only xgboost and glms (trained on the 512-D embeddings generated with tensorflow) are supported, however in a future update we will add functionality to allow arbitrary sentiment scoring models.

Simple Example

# Load the package
require(sentiment.ai)
require(SentimentAnalysis)
require(sentimentr)

# Only if it's your first ever time
# sentiment.ai.install()

# Initiate the model
# This will create the sentiment.ai.embed model
# Do this so it can be reused without recompiling - especially on GPU!
init_sentiment.ai()

text <- c(
    "What a great car. It stopped working after a week.",
    "Steve Irwin working to save endangered species",
    "Bob Ross teaching people how to paint",
    "I saw Adolf Hitler on my vacation in Argentina...",
    "the resturant served human flesh",
    "the resturant is my favorite!",
    "the resturant is my favourite!",
    "this restront is my FAVRIT innit!",
    "the resturant was my absolute favorite until they gave me food poisoning",
    "This fantastic app freezes all the time!",
    "I learned so much on my trip to Hiroshima museum last year!",
    "What happened to the people of Hiroshima in 1945",
    "I had a blast on my trip to Nagasaki",
    "The blast in Nagasaki",
    "I love watching scary horror movies",
    "This package offers so much more nuance to sentiment analysis!",
     "you remind me of the babe. What babe? The babe with the power! What power? The power of voodoo. Who do? You do. Do what? Remind me of the babe!"
)

# sentiment.ai
sentiment.ai.score <- sentiment_score(text)

# From Sentiment Analysis
sentimentAnalysis.score <- analyzeSentiment(text)$SentimentQDAP

# From sentimentr
sentimentr.score <- sentiment_by(get_sentences(text), 1:length(text))$ave_sentiment


example <- data.table(target = text, 
                      sentiment.ai = sentiment.ai.score,
                      sentimentAnalysis = sentimentAnalysis.score,
                      sentimentr = sentimentr.score)
targetsentiment.aisentimentAnalysissentimentr
What a great car. It stopped working after a week.-0.70.40.09
Steve Irwin working to save endangered species0.270.17-0.09
Bob Ross teaching people how to paint0.2800
I saw Adolf Hitler on my vacation in Argentina…-0.2900.27
the resturant served human flesh-0.320.250
the resturant is my favorite!0.80.50.34
the resturant is my favourite!0.7800
this restront is my FAVRIT innit!0.6300
the resturant was my absolute favorite until they gave me food poisoning-0.3600.12
This fantastic app freezes all the time!-0.410.250.13
I learned so much on my trip to Hiroshima museum last year!0.6400
What happened to the people of Hiroshima in 1945-0.5800
I had a blast on my trip to Nagasaki0.73-0.33-0.13
The blast in Nagasaki-0.51-0.5-0.2
I love watching scary horror movies0.540-0.31
This package offers so much more nuance to sentiment analysis!0.7400
you remind me of the babe. What babe? The babe with the power! What power? The power of voodoo. Who do? You do. Do what? Remind me of the babe!0.550.3-0.05

Benchmarks

So, what impact does more robust detection and some broader context have? To test it, in real-world scenarios, we use two datasets/use cases:

  1. classifying whether review text from Glassdoor.com is from a pro or a con

  2. the popular airline tweet sentiment dataset.

We use the default settings for sentimentr, the QDAP dictionary in sentimentAnalysis, and en.large in sentiment.ai. We prefer the use of Kappa to validate classification as it’s a less forgiving metric than F1 scores. In both benchmarks sentiment.ai comes out on top by a decent margin!

Note that our testing and tuning was one using comments written in English.

Glassdoor

Applied example, estimating whether the text from a glassdoor.com review is positive or negative. The validation set used here is the same data KFI used in our 2020 SIOP workshop

Note: As a part of KFI’s core purpose, sentiment.ai‘s scoring models were tuned with extra work-related data, hence this is tilted in our favor!

Airline Tweets

Taken from the airline tweet dataset from Kaggle. Classification is positive vs negative (neutral was omitted to remove concerns about cutoff values).

Note: Azure Cognitive Services tune their sentiment model on product reviews, as such this is tilted in favor of Azure!

Fierce. It looks like we can be pretty confident that sentiment.ai is a pretty fab alternative to existing packages! Note that over time our sentiment scoring models will get better and better!




Installation & Setup

New installation

After installing sentiment.ai from CRAN, you will need to make sure you have a compatible python environment for tensorflow and tensorflow-text. As this can be a cumbersome experience, we included a convenience function to install that for you:

install_sentiment.ai()

This only needs to be run the first time you install the package. If you’re feeling adventurous, you can modify the environment it will create with the following paramaters:

# Just leave this as default unless you have a good reason to change it. 
# This is quite dependent on specific versions of python moduules
install_sentiment.ai()

Assuming you’re using RStudio, it can be helpful to go to tools > global options > python > python interpreter and set your new tensorflow-ready environment as the default interpreter. There’salso an option for automatically setting project-level environments,

Note for GPU installations: you’ll need to make sure you have a compatible version of CUDA installed. (Here is a helpful guide to pick your CUDA version)[https://www.tensorflow.org/install/source#gpu]




Initialize

You’ll probably want to initialize the language embedding model first if you want to 1) not use the default models, and 2) want to re-apply sentiment scoring without the overhead of preparing the embedding model in memory. Pre-initializing with init_sentiment.ai() is useful for making downstream sentiment scoring and matching run smoothly, especially on GPU.

Technically init_sentiment.ai() will give access to a function called sentiment.ai_embed$f() which uses the pre-trained tensorflow model to embed a vector of text. Power users, the curious, and the damned may want to look in (Helper Functions for an example of this magic)[#power-move].

init_sentiment.ai() has the following optional paramaters:

model

The default Universal Sentence Encoder models available are:

Example:

# NOTEL In most cases all you need is this:!!
init_sentiment.ai()

# To change the default behavior:
# e.g. to initialise for use on multi lingual comments:
# leave envname alone unless you have a good reason to change it! 
init_sentiment.ai(model = "multi.large")

envname

default is envname = "r-sentiment-ai" – only change this if you set a different environment in install_sentiment.ai()


Sentiment Analysis

sentiment_score()

For most use cases, sentiment_score() is the function you’ll use.

Returns a vector of sentiment scores. The scores are a rescaled probability of being positive (i.e 0 to 1 scaled as -1 to 1). These are calculated with a secondary scoring model which, by default is from xgboost (a simple GLM is available if for some reason xgboost doesn’t work for you!).

The paramaters sentiment_score() takes are:

Note that if init_sentiment.ai() has not been called before, the function will try to recover by calling it into the default environment. If you’re using CUDA/GPU acceleration, you’ll see a lot of work happening in the console as the model is compiled on the GPU.

For example:

my_comments <- c("Will you marry me?", "Oh, you're breaking up with me...")

# for English, default is fine
sentiment_score(my_comments)

sentiment_match()

Sometimes you want to classify comments too. For that, we added sentiment_match() which takes a list of positive and negative terms/phrases and returs a dataframe like so:

TextSentiment scorePhrase matchedClass of phrase matchedSimilarity to phrase
“good stuff”0.78“something good”positive.30

While there are default lists of positive and negative phrases, you can overwrite them with your own. In this way you can quickly make inferences about the class of comments from your specific domain. The sentiment score is the same as calling sentiment_score() but you also get the most similar phrase, the category of that phrase, and the cosine similarity to the closest phrase.

For example:

my_comments <- c("Will you marry me?", "Oh, you're breaking up with me...")

my_positives <- c("excited", "loving", "content", "happy")
my_negatives <- c("lame", "lonely", "sad", "angry")

my_categories <- list(positive = my_positives, negative = my_negatives)

result <- sentiment_match(x = my_comments, phrases = my_categories)

print(result)


##                                 text  sentiment phrase    class similarity
## 1:                Will you marry me?  0.5393031 loving positive  0.1723714
## 2: Oh, you're breaking up with me... -0.6585207    sad negative  0.1512707

Note Cosine similarity is relative here & longer text will tend to have lower overall similarity to a specific phrase!

Note 2 You can also be tricky and pass in a list of arbitrary themes rather than just positive and negative – in this way sentiment.ai can do arbitrary category matching for you!


Matrix Analysis

cosine()

A light equivilent of text2vec::sim2 that gives pairwise cosine similarity between each row of a matrix.

x1 <- matrix(rnorm(4 * 6), 
             ncol = 6,
             dimnames = list(c("a", "b", "c", "d"), 1:6))

x2 <- matrix(rnorm(4 * 6), 
             ncol = 6,
             dimnames = list(c("w", "x", "y", "z"), 1:6 ))

# to check that it's the same result
# all.equal(cosine(x1, x2), text2vec::sim2(x1, x2))

cosine(x1, x2)

##            w          x           y           z
## a -0.1776981 -0.1743733  0.04111033  0.88174300
## b -0.5699782  0.6938432  0.21893192  0.05391341
## c  0.5076042  0.0169079  0.04563749  0.38885764
## d  0.2050958 -0.7644069 -0.74160285 -0.13175003

cosine_match()

This is a helper function to take two matrices, compute their cosine similarity, and give a pairwise ranked table. For example:

cosine_match(target = x1, reference = x2)

##     target reference  similarity rank
##  1:      a         w -0.17769814    4
##  2:      b         w -0.56997820    4
##  3:      c         w  0.50760421    1
##  4:      d         w  0.20509577    1
##  5:      a         x -0.17437331    3
##  6:      b         x  0.69384318    1
##  7:      c         x  0.01690790    4
##  8:      d         x -0.76440688    4
##  9:      a         y  0.04111033    2
## 10:      b         y  0.21893192    2
## 11:      c         y  0.04563749    3
## 12:      d         y -0.74160285    3
## 13:      a         z  0.88174300    1
## 14:      b         z  0.05391341    3
## 15:      c         z  0.38885764    2
## 16:      d         z -0.13175003    2

If you filter that to only the rows where rank is 1, you’ll have a table of the top matches between target and reference.

embed_text()

As an added bonus of manually calling init_sentiment.ai() you can call embed_text() to turn a vector of text into a numeric embedding matrix (e.g if you want to cluster comments).

For example

# Note, there's some weird behavior when you do this with a single string! 
test_target <- c("dogs", "cat", "IT", "computer")
test_ref    <- c("animals", "technology")

# Work it!
target_mx <- embed_text(test_target)
ref_mx    <- embed_text(test_ref)
ref_mx[1:2, 1:4]

##                    [,1]       [,2]       [,3]          [,4]
## animals    -0.042174224 0.01914462 0.07767479 -0.0362021662
## technology  0.005067721 0.02222563 0.06019276  0.0006833972

# And slay.
result <- cosine_match(target_mx, ref_mx)
result[rank==1] # filtered to top match (with data.table's [] syntax)

##      target  reference similarity rank
## 1:     dogs    animals  0.8034535    1
## 2:      cat    animals  0.6122806    1
## 3:       IT technology  0.4383662    1
## 4: computer technology  0.6668407    1



Roadmap

We want to continue making sentiment analysis easier and better, to that end future releases will include:




Contribute a Model!

Think you can make a better sentiment scoring model? Seeing as I’m far from the sharpest lighbulb out there, you probably can out do me! Simply train a model on text embeddings and add it to the models folder of this repo. Currently we only have support for xgb and glm models (xgb worked really well with minimal faff, and GLMs, saved as simple parameter weights, are super light weight). We will figure out a way to allow custom models, likely with a custom predict script in each model folder, but that’s a tomorrow-problem for now.

The models directory contains models used to derrive sentiment after the neural nets have handles the embedding bit.

Directory structure is: model_type/version/< embedding name="">.< ext>

For example: models/xgb/1.0/en.large.xgb is the default scoring model in sentiment_score(x) which is the same as a user passing sentiment_score(model="en.large", scoring="xgb", scoring_version="1.0").When called, sentiment.ai will pull the scoring model from github if it hasn’t done so already (default is installed during install_sentiment.ai()) and apply that to the embedded text.

NOTE

If you’d like to contribute a new/better/context specific model, use the same folder structure If it’s not a glm or XGB, let us know so we can make it work in find_sentiment_probs()! For non-xgb/glm models, please give an example R script of applying it to a matrix of embedded text so we can add support for it.

We only ask that numeric version (e.g xgb/2.0/…) names be left for official/default models. For community models, the version can be a descriptor (so long as it’s a valid file name and URL!) e.g: xgb/jimbojones_imdb1/en.large.xgb This way we can keep it easy for less engaged people who want a package that “just works” while accommodating power users that want custom models

If you have a general purpose model (i.e. trained on a variety of sources, not context specific) that out-performs the default ones, get in touch, we’ll test some extra benchmarks, and if it’s “just better” we’ll add it to become the new official/default (obviously giving you credit!)

will add support to pass through custom github urls if you don’t feel like sharing. When that’s done you’d pass in: repo_url = “https://github.com/< your name="">/< repo name="">/raw/< branch>/models” in the … of sentiment_score() or sentiment_match().


To leave a comment for the author, please follow the link and comment on their blog: Happy Little Scripts.

R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.