MeaningCloud Europe is now part of Reddit

Reddit & MeaningCloud

I’m thrilled to announce that MeaningCloud Europe (formerly Daedalus), the company I started in 1998, is now a part of Reddit. You can read the full announcement of this acquisition on Reddit’s blog.

You all know Reddit, of course, the online platform hosting more than 100,000 communities where people can discuss and chat about their interests, hobbies, and passions.

And you know MeaningCloud, the Language Technology company that competes (modestly but proudly) with IBM Watson, Google Cloud, Amazon Comprehend, and Microsoft Cognitive Services. Some of you may remember that we started offering public text analytics SaaS services under the name Textalytics in 2013, more or less simultaneously with the commercial launch of IBM Watson APIs.

As geeks, working on Content Understanding as part of the fantastic Reddit’s machine learning (ML) team, around the largest internet corpus of conversational content, is a dream come true. There is no need to explain the excitement of our team about the new challenges we face as part of Reddit.

At this point, some of you, users, clients, or friends who follow us, may be wondering what happens now with our current businesses. I’m happy to announce that, as part of the deal, we have transferred our clients to our shareholder Sngular. Sngular (Singular People, S.A.) is a global provider of IT services, a Spanish public company with 1,300 employees. At MeaningCloud, we are proud to have participated as founding partners of Sngular in 2015. In particular, Sngular is taking over:

Sngular

  • Our wholly-owned subsidiaries, MeaningCloud LLC and Konplik Health Inc. The first one will keep running our text analytics platform and servicing our clients and users worldwide (including 60,000 users of our free tier). Konplik will continue providing AI services to the health and pharma industry.
  • Our stake at Expoune Inc., a services company built around exploiting MeaningCloud’s technology in different settings.
  • Stilus, our automatic spell, grammar, and style assistant for Spanish, with 250,000 registered users, the Spanish checker preferred by media companies and professionals (translators, copyeditors, journalists, and lovers of the Spanish language).
  • Our clients in different industries (consulting, finance, infrastructure, telecommunications, etc.) for which we develop specific embedded solutions (SaaS or on-prem) on top of our technology.

So, we are leaving our business in the best hands, under the direction of Cesar Camargo (Sngular CEO), Alma Miller (CEO of Sngular in the United States), and José Luis Calvo (head of Artificial Intelligence). It’s your baby now!

You may have your preferred case among pivoting companies: Netflix, IBM, Western Union, Android, Nintendo… They made substantial changes in their core activities or strategy over the years. What about us? Competing in hi-tech is (extremely) tough. Starting an Artificial Intelligence company in Spain in 1998 was evidence of madness (shared with my colleagues Juan R. Velasco and Luis Magdalena). And taking it to here (along with Antonio Laorden, José L. Martínez and Julio Villena) was against any odds. On the way, we pivoted our business several times. For example, in the last seven years, we took part in the birth of Sngular (following the vision of Jose Luis Vallejo), created two new US-based companies, and took a stake in Expoune (led by Robert Wescott).

On the technology side, however, we have followed a much more consistent path based on three principles: technological agnosticism, problem-solving mindset, and client orientation. For example, Stilus was our first Natural Language Processing (NLP) product. The newspaper El País was our first corporate client in the news industry in 2004. And, 18 years later, we keep maintaining Stilus, under the same license contract, despite the changes in technology, product, people, and content platforms.

Blog Reddit

Today, we pivot again to be part of a large and fast-growing company in the social media arena. And we face this challenge with all the experience and assets accumulated over the years. Time is the best judge of our actions. However, I’m sure about the great value that we will be contributing to Reddit in the coming years, configuring one of the leading organizations in the exploitation of NLP  technologies, beyond the current hype around ML and Artificial Intelligence (AI) in general.

For me, all this means turning over a new leaf. Having got my Ph.D. in AI in the late 80s, 30 years as a university professor, 24 years since I started up my first company… I meet all the requirements to be the senior of “intern snoos” at Reddit. I’m committed to doing my part, side by side with my incredible team, under the direction of Jack Hanlon and Rosa Català, to contribute, in my poor power, to Reddit’s growth and international expansion. It sounds like a plan, right?

Thank you, and good luck to all the people (past employees, investors, clients, users, and friends) who have supported us along the road. And let’s meet on Reddit!

All the best,

Jose C. Gonzalez

 


Corporate Reputation Analysis at Scale

The rationale for Corporate Reputation Analysis Automation

Reputation

We live in an age where news stories are no longer a primary, but an added concern for businesses trying to build and maintain a strong reputation. Individual experiences are reaching a global audience in a matter of minutes, thanks to the internet, which has made way for an immense volume of spontaneous and real-time information. There is no doubt that companies have to navigate the voices of traditional and contemporary media sources, both of whom contribute significantly to their social standing. Reputation crises can spark at any moment, and traditional reputational audits can not help to mitigate them. Consequently, it is more important than ever before to keep track of reputation in real-time.

The MeaningCloud Corporate Reputation Analysis API is a new service that enables companies to take advantage of social networking platforms, forums, blogs, surveys, and news sources in a bid to do just that. While we have been running a service in this field since 2011, our new API employs a distinct approach to tackle the task of Corporate Reputation analysis.

Reputation management fundamentals

The MeaningCloud Reputation API uses a categorization scheme based on the work by Charles Fombrun and Cees van Riel. In 1999, they founded The Reputation Institute (now, The RepTrak Company™), the world’s leading reputation data and insights company. In collaboration with Harris Interactive, The Reputation Institute developed Reputation Quotient (RQ) in 1999, replaced in 2005 by the RepTrak® model.

The RepTrak Company™ has become highly esteemed on a global scale for its publication of reports on corporations’ reputation (as well as countries and cities). Furthermore, it has developed models that have provided companies with the autonomy to qualify their reputation in a meaningful way. It is important to note that other well-known frameworks do exist. The MERCO Indexes, for instance, are particularly renowned in Spain, Portugal, Italy, and across Latin America for their “multi-stakeholder methodology, composed of six evaluations and more than twenty information sources”. MERCO has been analyzing reputation since the year 2000.

To analyze Corporate Reputation in a piece of text, we follow the seven dimensions proposed in the foundational work of this field, globally considered to be the most important drivers of reputation:

  1. Citizenship:
    Both a company that actively strives and one who is idle in their efforts to be environmentally responsible, ensuring they support good causes and positively influence society, are equally susceptible to a change in their reputational outlook.
  2. Governance:
    A company needs to be open and transparent, ethical and fair in order to boost its reputation.
  3. Innovation:
    Innovative companies are market leaders and adapt quickly to change generally have a better social standing.
  4. Leadership:
    Suppose a company has inspiring/motivating leaders who display personal integrity, competence and knowledge, effective communication, and a clear vision for the future. They are likely to have a positive effect on the company’s repute.
  5. Performance:
    A measure of determining how profitable a company is, whether they exceed expectations and have solid prospects for the future. All of which are qualities of a reputable company.
  6. Products and Services:
    High-quality products/services that meet customer needs and are fully backed by the company undoubtedly contribute to a company’s social status.
  7. Workplace:
    A company must reward employees fairly, show genuine concern for them, and respect diversity to maintain a good reputation.

Our new Corporate Reputation API

The MeaningCloudReputation API is now available in 57 languages. The core processing is carried out in English. For other languages, the source text is translated first with the MeaningCloud Machine Translation API. For example, a user can throw an API request with a text in German. He/she can specify the source language (if known) as a parameter in the call or select “autodetect” to let our algorithms discover the source language, translate it into English (if needed), and carry out the reputation analysis on the translated text.

The MeaningCloud Corporate Reputation API goes beyond the automatic discovery of a mention related to a particular dimension in a Corporate Reputation model:

  1. It first detects the language of the text under analysis and translates it into English if needed.
  2. It disambiguates the mentions of entities in the text. Users can extend the (vast) resources updated daily in the platform with new companies via user dictionaries.
  3. It discriminates the reputation dimensions evoked in the text.
  4. It analyzes each dimension’s polarity (positive, negative, neutral, or non-existent) in a reputational context.
  5. It attaches the pair dimension/polarity to one of the entities mentioned in the text or to a declared default entity.

Reputation Management

Tools for PR, communication, marketing, and social listening companies

MeaningCloud is just a tech company specialized in Natural Language Processing. We do not collect information over the internet to assess the reputation of companies. Instead, we build the technology that permits third parties to develop and leverage NLP technology to provide actionable insights into:

  • The overall perception of the company.
  • The impact that news and social media content have on the company’s reputation.
  • The areas of the company requiring special attention.
  • The areas of the company making a good impression.
  • Third companies that are mentioned jointly and how their respective reputations compare.

The limits of the technology

As always with NLP solutions, and despite our continuous efforts, this API cannot be entirely free from errors or biases coming from different sources:

  • Our interpretation of the dimensions of reputation, that we know through the widely available literature on this subject.
  • Limitations in our algorithms for linguistic analysis and translation models.
  • Lack of adaptation to a particular domain or industry. Our general-purpose API can be improved to interpret reputational aspects more precisely when analyzing companies in a particular industry: utilities, finance, retail, telecom, healthcare, etc.

Our Corporate Reputation API derives from:

  • Our experience in customer feedback analysis, using the same basis as our Voice of the Customer analysis models.
  • The version of the API for Spanish, which has been public since 2012 (at textalytics.com until 2015 and at meaningcloud.com since then.)

Free Corporate Reputation Analysis

Our API is now published in beta version. For now, the pricing is the same as for other premium APIs: one request (or credit) is charged per 125 words or fraction. Two credits if translation is required. However, temporarily, there is no need to pay a flat fee for usage, as is the case for our Voice of the Customer/Employee APIs.

Follow this link to learn more about our Corporate Reputation API.

Furthermore, remember that you can test the API extensively, analyzing up to 20,000 texts for free per month, just by registering at meaningcloud.com.

Disclaimer

To be completely transparent about our credentials, we have to make it clear that:

Corporate Reputation ReviewReferences

An excellent source of information about this field is the journal Corporate Reputation Review. Launched in 1997, it publishes empirical and conceptual research on reputation management and closely related fields, such as strategic/corporate communication, corporate social responsibility (CSR) communication, corporate identity, and organizational identity.

Fombrun, C. and Shanley, M., 1990. What’s in a name? Reputation building and corporate strategy. Academy of management Journal, 33(2), pp.233-258.

Fombrun, C. and Van Riel, C., 1997. The reputational landscape. Corporate reputation review, pp.1-16.

Fombrun, C.J., Gardberg, N.A. and Sever, J.M., 2000. The Reputation Quotient SM: A multi-stakeholder measure of corporate reputation. Journal of brand management, 7(4), pp.241-255.

Fombrun, C.J., Van Riel, C.B. and Van Riel, C., 2004. Fame & fortune: How successful companies build winning reputations. FT press.

Ponzi, L.J., Fombrun, C.J. and Gardberg, N.A., 2011. RepTrak™ pulse: Conceptualizing and validating a short-form measure of corporate reputation. Corporate reputation review, 14(1), pp.15-35.

Fombrun, C.J., Ponzi, L.J. and Newburry, W., 2015. Stakeholder tracking and analysis: The RepTrak® system for measuring corporate reputation. Corporate reputation review, 18(1), pp.3-24.

Pallarés Renau, M. y López Font, L., 2017. Merco y RepTrak Pulse: Comparación cualitativa de atributos, variables y públicos, Icono 14, volumen 15 (2), pp. 190-219.

 

Janine Garcia, Nadine Shallow, Maria Jose Garcia, Concepcion Polo, and Jose Gonzalez


Full IAB taxonomy for content classification now available

The surge of content marketing, based on the creation and distribution of valuable content, is a reality. The objective is no longer solely to advertise but to engage and offer valuable experiences to the market as well. In this context, precision in content classification is becoming essential. And that’s why we implemented a new automatic content classifier according to the full IAB taxonomy.

Therefore, since including tier 3 of the content taxonomy to the categorization categories of our IAB model last year, we have chosen to complement it by providing the complete ontology. In line with the Content Taxonomy, published in 2017 and updated in 2020 by the IAB Tech Lab (Interactive Advertising Bureau), this classification includes the remaining 60 categories that make up tier 4.  A total categorization of over 698 categories, hierarchized into four tiers, is thus, offered. We have kept the unique identifier that IAB assigns to each one of its categories, as well as their name, through which its parent categories can be found.

Continue reading


A Quick Introduction to Brand Monitoring and Brand Protection

[EDITOR’S NOTE: This is a guest post by David Bitton, Product Manager at Webhose.io]

Organizations should never underestimate the power of their brand. What starts out as a name, logo, vision, mission statement, website, and perhaps a few employees start to form an organization’s identity. As an organization grows, all of these key parts evolve over time to help customers identify the brand.

But a brand is more than its identity. A brand should evoke emotion from its customers – ideally a positive one – creating brand loyalty and repeat purchases of its goods or products. Loyal customers also refer the brand to friends, family, and acquaintances.

In today’s digital age, building and maintaining a strong brand is so important that brand monitoring plays a crucial role in organizations’ marketing strategy. However, with the rise of the dark web, brand monitoring has evolved to now include Digital Risk Protection. DRP protects the brand’s digital assets from various malicious actors intent on causing the brand and its reputation significant damage.

Continue reading


MeaningCloud brings multilingual text analytics to 57 languages!

It’s time to celebrate! We are rolling an update to our APIs that allows analyzing texts in fifty-seven languages. Our dream of multilingual text analytics based on our deep semantic approach has come true.

Until now, to analyze texts in different languages, we needed to maintain a model per language. This gives good results, although it was hard and expensive. We can do better.

Zero to near sixty in no time

We have integrated our APIs with deep neural network technology to translate all these languages into English. Thanks to this, our users can analyze texts in many languages maintaining only one model. And no action is required. All the new languages will appear in your test console and are available in the APIs.

The complete list of languages includes Chinese, Hindi, Arabic, Russian, Japanese, Turkish, German, and many others that our customers have requested frequently. This adds to our current offering for English, Spanish, French, Italian, Portuguese, and the languages we were covering partially (in some APIs).

Multilingual Text Analytics

Continue reading


Machine Learning for NLP/Text Analytics, beyond Machine Learning

In the field of text analytics, aside from the development of categorization models, the application of machine learning (and more specifically, deep learning) has proved to be very helpful for supporting our teams in the process of building/improving rule-based models.

This post analyzes some of the applications of machine/deep learning for NLP tasks, beyond machine/deep learning itself, that are used to approach different scenarios in projects for our customers.

Continue reading


Accuracy measures in Sentiment Analysis: the Precision of MeaningCloud’s Technology

Accuracy Measures of Commercial Sentiment Analysis APIs

Our clients frequently ask, “what’s the precision of MeaningCloud technology?” How does it compare with other commercial competitors and with state-of-the-art technology? And they demand precise numbers.

That’s not an easy question to answer. Even when there are milliards of research studies on this issue. For the sake of simplicity, let’s concentrate on the well-studied scenario of accuracy measures in Sentiment Analysis. Continue reading


New Excel 365 add-in for Text Analytics!

Our new Excel 365 add-in has finally arrived!

Excel is the preferred tool for many MeaningCloud users. They access MeaningCloud APIs directly from Excel with our add-in. In the last months, we have received a lot of inquiries about Mac support. So, we partnered with Microsoft to build a new multiplatform version.

 

Installation

Installing it is a breeze on all platforms. The new add-in is available in Microsoft AppSource:

https://appsource.microsoft.com/en-us/product/office/WA200002421

Click on Get it Now and follow the instructions.

 

Configuration

You only need your API key to use MeaningCloud. Paste it in the License Key field and you’re ready to start analyzing.

Don’t have one? Create an account for free – no payment method required.

Configuring the MeaningCloud add-in

 

Usage

You can use the APIs directly from the ribbon:

MeaningCloud add-in ribbon

The user interface page describes the different buttons. Paste your texts in the spreadsheet, select the tool in the ribbon, review the parameters and click in Analyze:
Sentiment analysis with MeaningCloud
Take a look at the documentation for more information about add-in usage.

 

But I don’t use Office 365!

No worries. If you use another Excel version, we still offer the previous add-in version. If you don’t use Microsoft Excel at all, you can use our Google Spreadsheets add-on.

 

Questions?

If you have any questions or issues, we will be glad to hear from you. Drop us a line at support@meaningcloud.com and tell us about your experience.

Language Technology vs. Disinformation

Reuters Institute Digital News Report 2020

Following the 2016 US presidential election, many have expressed concern about the effects of false stories (“fake news”), circulated largely through social media. Research from Oxford University’s Reuters Institute for the Study of Journalism has found a long and steady decline in trust in traditional media. (See Reuters Institute Digital News Report 2020). This declining trust coincides with the uprise of social media as a main source of information. In 2020, social media was a source of news for 48 percent of the public, up from 27 percent in 2013, according to the Reuters Institute.

Continue reading


IAB Taxonomy Level 3 now available in our Deep Categorization API

IAB - Interactive Advertising BureauDigital marketing is becoming a fundamental pillar, by leaps and bounds, in the business plans of practically every business model. Methods are being refined and the search for the connection between brand and user is expected to become increasingly more precise: a related advertisement is no longer sufficient, now the advertisement must appear at the right time and in the right place. This is where categorization proves to be an exceedingly useful tool.

That is why, at MeaningCloud, we have improved our IAB categorization model in English, that is integrated in our Deep Categorization API:

  • Adding a third level of content taxonomy to the hierarchy of categories (IAB Taxonomy Level 3).
  • Improving the precision of pre-existing categories.
  • Including the unique identifiers defined by IAB itself for each of the categories.

Continue reading