Language Tech Market News
52.5K views | +33 today
Follow
Language Tech Market News
The Home of Multilingual Intelligence
Curated by LT-Innovate
Your new post is loading...
Your new post is loading...
Scooped by LT-Innovate
Scoop.it!

Facebook's 'Rosetta' Extracts Text from Images to Detect Sentiment

Facebook's 'Rosetta' Extracts Text from Images to Detect Sentiment | Language Tech Market News | Scoop.it

Rosetta starts by detecting rectangular regions in images that potentially contain text. It then uses a convolutional NN to recognize and transcribe what's written - non-English words and non-Latin alphabets such as Arabic and Hindi. The system is trained on a mix of human- and machine-annotated public images.

more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

Email Text Analytics to Gauge Corporate Morale

Email Text Analytics to Gauge Corporate Morale | Language Tech Market News | Scoop.it
In an ideal world, employees would be honest with their bosses, and come clean about all the problems they observe at work. But in the real world, many employees worry that the messenger will be shot; their worst fears stay bottled up. Text analytics might allow firms to gain insights from their employees while intruding only minimally on their privacy. The lesson: Figure out the truth about how the workforce is feeling not by eavesdropping on the substance of what employees say, but by examining how they are saying it.
more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

@Sinequa Recognized as a Leader in Gartner 2018 Magic Quadrant for Insight Engines

more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

Salesforce #NLProc Model Performs 10 Language Tasks Simultaneously

Salesforce #NLProc Model Performs 10 Language Tasks Simultaneously | Language Tech Market News | Scoop.it

DecaNLP puts the MQAN through linguistic tests, including question-answering (in which the model receives a question and a context that contains the information necessary to arrive at an answer), and machine translation (which has the model translate an input document from one language to another). There’s a document summarization test, a natural language inference test, a sentiment analysis test, a semantic role labeling test, a relation extraction test, a goal-oriented dialog test, a query generation test, and a pronoun resolution test.

more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

Curse Rate Doubles After 4pm: Phone Study Reveals Spoken Content Trends Across America

Curse Rate Doubles After 4pm: Phone Study Reveals Spoken Content Trends Across America | Language Tech Market News | Scoop.it

Among other findings in this Marchex Institute report, cursing increases as the day progresses. On average, callers cursed 1.1 to 1.3 times per call across the board. However, cursing becomes more common as the day progresses. Profanity is least likely to be used before 4:00 p.m., but then the curse rate on calls doubles after 6:00 p.m., jumping from three percent to six percent. The data also shows that instances of cursing are nuanced. For example, s**t is the most common curse word, but can be used differently depending on the tone of the conversation.

more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

Deep Learning for Sentiment Analysis : A Survey

Deep learning has emerged as a powerful machine learning technique that learns multiple layers of representations or features of the data and produces state-of-the-art prediction results. Along with the success of deep learning in many other application domains, deep learning is also popularly used in sentiment analysis in recent years. This paper first gives an overview of deep learning and then provides a comprehensive survey of its current applications in sentiment analysis.
LT-Innovate's insight:

Scientific paper surveying the field

more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

#NLProc Unlocks New Frontier in Corporate Earnings Sentiment

#NLProc Unlocks New Frontier in Corporate Earnings Sentiment | Language Tech Market News | Scoop.it

According to Seagate Technology's new report, Natural Language Processing – Part 1: Primer, CEOs using more words per sentence or increasing the number of polysyllabic words in their earnings calls, can be telltale signs of future earnings and share price declines. Perhaps it should come as little surprise, then, that Seagate has lost approximately 35% of its market between April 26, 2017 and September 7, 2017.

LT-Innovate's insight:

Longwinded language in earnings calls as tell-tale signs of difficulties ahead. Nice to have some proof, though, of what humans sense about "woffle".

more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

Training Sentiment Analysis Using Crowdsourcing 

Training Sentiment Analysis Using Crowdsourcing  | Language Tech Market News | Scoop.it

I spoke with Charly Walther, VP of Product and Growth at Gengo.ai, the crowdsourced AI training arm of Gengo.com – a web-based human translation platform headquartered in Tokyo. I asked Charly about specific circumstances where crowdsourced human perception is necessary to train a sentiment system well.

more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

Cogito Uses Spoken Sentiment Analysis to Improve Call-centre Service

Cogito Uses Spoken Sentiment Analysis to Improve Call-centre Service | Language Tech Market News | Scoop.it

Cogito has announced a $37 million Series C investment. The company has raised over $64 million since it emerged from the MIT Human Dynamics Lab back in 2007 trying to use the artificial intelligence technology available at the time to understand sentiment and apply it in a business context.
While it took some time for the technology to catch up with the vision, and find the right use case, company CEO and founder Joshua Feast says today they are helping customer service representatives understand the sentiment and emotional context of the person on the line and give them behavioral cues on how to proceed.

more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

@CBInsights: Linguistic Analysis of Automotive Earnings Calls 

@CBInsights: Linguistic Analysis of Automotive Earnings Calls  | Language Tech Market News | Scoop.it

Our machine learning algorithm analyzed 28 quarters of earnings transcripts from Ford, General Motors, Daimler, and Tesla to bring you an in-depth look at their strategies and forward-looking moves.
- How Mobility CEOs compare to each other across 4 linguistic algorithms
- Which automakers talk most about the competition
- The strategies of Tesla, General Motors, Ford, and Daimler
- A case study of Hertz & Avis

more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

Reuter's News Tracer at Data Summit 2018

Reuter's News Tracer at Data Summit 2018 | Language Tech Market News | Scoop.it

Reuters is introducing a new tool called News Tracer. It is a capability that applies AI in journalism to find events breaking on Twitter. It assigns them a newsworthiness score so people can focus on the events that are important.

more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

European Central Bank Publishes #NLP-driven Monetary Policy Tone Index

European Central Bank Publishes #NLP-driven Monetary Policy Tone Index | Language Tech Market News | Scoop.it

The ECB presents the main results of applying NPL techniques to the Statements and the Q&A of the European Central Bank press conferences on monetary policy from 1999 to 2017. The first section (2.1) describes some of the 30 latent vectors extracted by the Dynamic Topic Model and we look at the evolution of these topics over time. Meanwhile the interconnectedness of the topics is explained in the third section (2.2). Finally, we show the sentiment of these topics, including the sentiment of the monetary policy stance in the last section (2.3).

LT-Innovate's insight:

Makes a nice comparison with voice data on sentiment picked up from  quarterly-results conversations recorded from major stock market companies.

more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

Values Hidden Inside Job Post Language

We used Textio to take a look at the most distinctive language used in the public job posts of ten prominent tech companies.* Each one showed distinct language patterns that showed up in statistically anomalous ways. The distinct phrases used by each company showed up in their jobs much more often than they did for other companies in the sample, and frequently way more often than average for the industry.

more...
No comment yet.
Scooped by LT-Innovate
Scoop.it!

Automated Sarcasm Detector Being Developed

Researchers have developed an algorithm that can automatically detect sarcasm much better than humans can, by combining linguitic processing with critical contextual information about the sender, etc.

LT-Innovate's insight:

Who would have thought?

more...
No comment yet.