Knowledge Models
730 views | +0 today
Follow
Your new post is loading...
Your new post is loading...
Scooped by Tom Vandermolen
Scoop.it!

A Pipeline for Distributed Topic and Sentiment Analysis of Tweets o...

Unstructured data is everywhere - in the form of posts, status updates, bloglets or news feeds in social media or in the form of customer interactions Call Cent (RT @being_bayesian: Slides for my talk 'Distributed Pipeline for Topic and Sentiment ...
Tom Vandermolen's insight:

Nice to see a pipeline of how the different platforms and methods link together, end-to-end.

more...
M. Edward (Ed) Borasky's comment, January 11, 2014 11:33 PM
Impressive ... until you consider the cost, anyhow. That kind of hardware and access to Gnip data isn't going to be cheap. There needs to be a big payoff somewhere, and frankly, I don't see it.
Scooped by Tom Vandermolen
Scoop.it!

Here's what we know about 'Cortex,' Twitter's new artificial intelligence group focused on understanding content - Business Insider

Here's what we know about 'Cortex,' Twitter's new artificial intelligence group focused on understanding content - Business Insider | Knowledge Models | Scoop.it
David Paul Morris/Bloomberg via Getty ImagesTwitter Interim CEO Jack DorseyTwitter is ramping up its artificial intelligence efforts, hunting for experts t...
Tom Vandermolen's insight:

Twitter joins the AI arms race.  The initial driver for their AI effort was to stretch the boundaries of our old definitions of intelligence, and--oh, wait.  No, sorry...it was porn and spam, the Two Horsemen of Technological Progress.

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

bat-country: an extendible, lightweight Python package for deep dreaming with Caffe and Convolutional Neural Networks - PyImageSearch

bat-country: an extendible, lightweight Python package for deep dreaming with Caffe and Convolutional Neural Networks - PyImageSearch | Knowledge Models | Scoop.it
bat-country is a lightweight, easy to use Python library that makes inceptionism and deep dreaming with Caffe and CNN painless with only 3 lines of code.
Tom Vandermolen's insight:

A refactorization of Google's deep dreaming module to make it more Python-friendly.  

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

Python For Sentiment Analysis Tutorial Introduction

A brief intro to Python, with the intent of using it for sentiment analysis/data mining/opinion mining. Sentdex.com Facebook.com/sentdex Twitter.com/sentdex.
Tom Vandermolen's insight:

Could come in handy.

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

Once More Without Feeling

Verbal identity consultants and copywriters investigate text analytics for marketing.
Tom Vandermolen's insight:

I understand they're tooting their own horn, and I actually think they have a good point...but why can't you do both topic and sentiment analysis?  Wouldn't the cross-checking help eliminate errors?

more...
No comment yet.
Rescooped by Tom Vandermolen from Applied linguistics and knowledge engineering
Scoop.it!

Corpora for Sentiment Analysis | Mark Cieliebak


Via Pascual Pérez-Paredes
Tom Vandermolen's insight:

Contains links to corpora used in their paper (which I need to check out).

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

Is sentiment analysis useful? | Our Social Times - Social Media Agency, Social Media Training

Is sentiment analysis useful? | Our Social Times - Social Media Agency, Social Media Training | Knowledge Models | Scoop.it
Nearly all social media monitoring tools offer sentiment analysis, but is it actually useful? The short answer is yes, but only if you get smart with the data.
more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

Universal Grammar- How Do You Back It? : linguistics

As I understand UG (admittedly through authors who don't agree with it), it's a non scientific theory made up as more of a philosophical thing by Chom...
Tom Vandermolen's insight:

Good discussion...I haven't read enough of the "con" side of the argument.

more...
No comment yet.
Rescooped by Tom Vandermolen from Text Analysis, Text Mining, Sentiment Analysis, Entity Extraction
Scoop.it!

Blogographic: Text Analytics to the Rescue

Blogographic: Text Analytics to the Rescue | Knowledge Models | Scoop.it
We spend significant amount of time sharing our thoughts and experiences by interacting with people. Be it about our organizations or about our opinion on upcoming elections or even about a new mob...

Via Ralph Poole
more...
Ralph Poole's curator insight, January 7, 2014 10:37 AM

This is a great graphic that explains the value of text analysis.

Scooped by Tom Vandermolen
Scoop.it!

The Sad State of Sentiment Analysis - Hausman Marketing Letter

The Sad State of Sentiment Analysis - Hausman Marketing Letter | Knowledge Models | Scoop.it
Sentiment analysis is problematic, yet its use is growing rapidly. Is there an alternative that is more accurate and insightful?
Tom Vandermolen's insight:

Some pretty good points. I'd be interested in seeing a results accuracy comparison between sentiment analysis of huge amounts of data and human evaluations of a random sampling.

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

On Chomsky and the Two Cultures of Statistical Learning

On Chomsky and the Two Cultures of Statistical Learning | Knowledge Models | Scoop.it
Tom Vandermolen's insight:

Interesting response by Peter Norvig to Noam Chomsky's criticisms of Statistical Linguistic analysis methods.

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

Natural Language Processing for the Working Programmer

Tom Vandermolen's insight:

This is a work in progress and it uses Haskell instead of more mainstream languages, but it seems to have a good reputation (wish I had a sentiment analysis program to measure that) and hey, it's free.

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

Hadoop Tutorial: Analyzing Sentiment Data

This video explores how to use Hadoop and the Hortonworks Data Platform to analyze sentiment data to understand how the public feels about a product launch. ...
more...
No comment yet.
Rescooped by Tom Vandermolen from Amazing Science
Scoop.it!

Deep Reinforcement Learning Machine Has Taught Itself to Play Chess at Higher Levels

Deep Reinforcement Learning Machine Has Taught Itself to Play Chess at Higher Levels | Knowledge Models | Scoop.it

"Chess, after all, is special; it requires creativity and advanced reasoning. No computer could match humans at chess." That was a likely argument before IBM surprised the world about computers playing chess. In 1997, Deep Blue's entry won the World Chess Champion, Garry Kasparov.


Matthew Lai records the rest: "In the ensuing two decades, both computer hardware and AI research advanced the state-of-art chess-playing computers to the point where even the best humans today have no realistic chance of defeating a modern chess engine running on a smartphone."


Now Lai has another surprise. His report on how a computer can teach itself chess—and not in the conventional way—is on arXiv. The title of the paper is "Giraffe: Using Deep Reinforcement Learning to Play Chess." Departing from the conventional method of teaching computers how to play chess by giving them hardcoded rules, this project set out to use machine learning to figure out how to play chess. Namely, he said that deep learning was applied to chess in his work. "We use deep networks to evaluate positions, decide which branches to search, and order moves."


As for other chess engines, Lai wrote, "almost all chess engines in existence today (and all of the top contenders) implement largely the same algorithms. They are all based on the idea of the fixed-depth minimax algorithm first developed by John von Neumann in 1928, and adapted for the problem of chess by Claude E. Shannon in 1950."


This Giraffe is a chess engine using self-play to discover all its domain-specific knowledge. "Minimal hand-crafted knowledge is given by the programmer," he said.


Results? Lai said ,"The results showed that the learned system performs at least comparably to the best expert-designed counterparts in existence today, many of which have been fine tuned over the course of decades."


OK, not at super-Grandmaster levels, but impressive enough. "With all our enhancements, Giraffe is able to play at the level of an FIDE [Fédération Internationale des Échecs, or World Chess Federation] International Master on a modern mainstream PC," he stated. "While that is still a long way away from the top engines today that play at super-Grandmaster levels, it is able to defeat many lower-tier engines, most of which search an order of magnitude faster."


Addressing the value of Lai's work in this paper, MIT Technology Review, stated that, "In a world first, an artificial intelligence machine plays chess by evaluating the board rather than using brute force to work out every possible move." Giraffe, said the review, taught itself to play chess by evaluating positions much more like humans.



Via Dr. Stefan Gruenwald
Tom Vandermolen's insight:

I wonder if this approach can be applied to a rule-bound environment like an OWL ontology, and used to learn how to automatically add new concepts and relationships to it?

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

» The LDA Buffet is Now Open; or, Latent Dirichlet Allocation for English Majors Matthew L. Jockers

» The LDA Buffet is Now Open; or, Latent Dirichlet Allocation for English Majors Matthew L. Jockers | Knowledge Models | Scoop.it
Tom Vandermolen's insight:

A fun, intuitive introduction to Topic Modeling using Latent Dirichlet Allocation (LDA).  This makes me wonder what kind of success anyone has had in using LDA to analyze Joyce's work.

more...
No comment yet.
Rescooped by Tom Vandermolen from Amazing Science
Scoop.it!

With deep learning and dimensionality reduction, we can visualize the entirety of Wikipedia?

With deep learning and dimensionality reduction, we can visualize the entirety of Wikipedia? | Knowledge Models | Scoop.it

Deep neural networks are an approach to machine learning that has revolutionized computer vision and speech recognition in the last few years, blowing the previous state of the art results out of the water. They’ve also brought promising results to many other areas, including language understanding and machine translation. Despite this, it remains challenging to understand what, exactly, these networks are doing.


Understanding neural networks is just scratching the surface, however, because understanding the network is fundamentally tied to understanding the data it operates on. The combination of neural networks and dimensionality reduction turns out to be a very interesting tool for visualizing high-dimensional data – a much more powerful tool than dimensionality reduction on its own.


Paragraph vectors, introduced by Le & Mikolov (2014), are vectors that represent chunks of text. Paragraph vectors come in a few variations but the simplest one, which we are using here, is basically some really nice features on top of a bag of words representation.


With word embeddings, we learn vectors in order to solve a language task involving the word. With paragraph vectors, we learn vectors in order to predict which words are in a paragraph.


Concretely, the neural network learns a low-dimensional approximation of word statistics for different paragraphs. In the hidden representation of this neural network, we get vectors representing each paragraph. These vectors have nice properties, in particular that similar paragraphs are close together.


Now, Google has some pretty awesome people. Andrew Dai, Quoc Le, and Greg Corrado decided to create paragraph vectors for some very interesting data sets. One of those was Wikipedia, creating a vector for every English Wikipedia article. The result is that we get a visualization of the entirety of Wikipedia. A map of Wikipedia. A large fraction of Wikipedia’s articles fall into a few broad topics: sports, music (songs and albums), films, species, and science.


Via Dr. Stefan Gruenwald
Tom Vandermolen's insight:

Another great machine learning/semantics tool.  We're getting closer, and it feels like all of these different techniques are homing in on *something* from different directions.

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

Chomsky’s “Universal Grammar” is incomplete

Chomsky’s “Universal Grammar” is incomplete | Knowledge Models | Scoop.it
Design constraints, rather than an innate knowledge of grammar, may explain similarities across languages
Tom Vandermolen's insight:

Interesting study..."Overall, though, this paper provides one of the clearest examples yet of where an important tendency in human language — a bias you would not expect to exist through mere random chance — can be explained by reference to universal principles of computation and information theory."

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

Sentiment Analysis Tools are Good – but not Perfect | Mark Cieliebak - Software Engineer and Researcher

Tom Vandermolen's insight:

Evaluated 9 commercial SA tools individually, then used a meta-classifier--finding that  a random forest model can improve results by up to 9%.

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

Big Content: The Unsung Hero of BI and Big Data

Big Content: The Unsung Hero of BI and Big Data | Knowledge Models | Scoop.it
Move over big data, it is big content that drives big value (Sentiment analysis could save the customer service experience.
Tom Vandermolen's insight:

I like the conceptual shift to Big Content.

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

List of 25+ Natural Language Processing APIs

List of 25+ Natural Language Processing APIs | Knowledge Models | Scoop.it
//
Natural Language Processing, or NLP, is a field of computer science, artificial intelligence, and linguistics concerned with the interactions between computers and human (natural) languages.
Here...
more...
Ralph Poole's curator insight, January 14, 2014 9:05 AM

This is a good list of of APIs that can be used for analysis and classification of textual content.  These tools are expecially relevant to sentiment analysis in which you are mining for opinion.

Scooped by Tom Vandermolen
Scoop.it!

Naming & Classifying: Text Analysis Vs. Text Analytics

Naming & Classifying: Text Analysis Vs. Text Analytics | Knowledge Models | Scoop.it
Analysts and marketers do a lot of naming and classifying -- This is a That -- in order to communicate distinctions among the many available products and technologies.
more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

Say What? Why Sentiment Analysis Continues to Fail Digital Marketers - CMSWire

Say What? Why Sentiment Analysis Continues to Fail Digital Marketers - CMSWire | Knowledge Models | Scoop.it
CMSWire
Say What? Why Sentiment Analysis Continues to Fail Digital Marketers
CMSWire
Take the term “sentiment analysis.” What does that really mean? According to Smith, it is the technology of making the large amount of data digestible.
Tom Vandermolen's insight:

Interesting that humans can only achieve ~80% accuracy in sentiment analysis...which kind of makes sense in light of how e-mail exchanges can get so inflammatory so quickly.  Context is everything, and without the right cues even we get it wrong.

more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

A simple way to explain Big Data to anyone - SuccessfulWorkplace

A simple way to explain Big Data to anyone - SuccessfulWorkplace | Knowledge Models | Scoop.it
A New York Times quiz on linguistics in the United States shows the key principles of Big Data analytics. (RT @avdingus: Got #BigData?
Tom Vandermolen's insight:

Nice...although technically it's a great example of Data Science and analytics more than Big Data (as the author points out at the end).

more...
No comment yet.
Rescooped by Tom Vandermolen from iMech
Scoop.it!

Deeply Moving: Deep Learning for Sentiment Analysis

Deeply Moving: Deep Learning for Sentiment Analysis | Knowledge Models | Scoop.it
This website provides a live demo for predicting the sentiment of movie reviews.

Via NikolaosKourakos
more...
No comment yet.
Scooped by Tom Vandermolen
Scoop.it!

What are good ways to get into Computational Linguistics?

What are good ways to get into Computational Linguistics? | Knowledge Models | Scoop.it
Answer (1 of 4): Having a BA in linguistics and being a web developer is one of the best positions for a self-learner!
Tom Vandermolen's insight:

Seems like some good roadmaps.  Still a bit confused on how much overlap or opposition exists between "traditional" linguistics and computational linguistics.

more...
No comment yet.