As we recently advanced, during these last few months we have been working on new functionality. We are planning to start releasing it over the next few months.
In the latest release of MeaningCloud we have included some of this functionality:
We have just published a new release of MeaningCloud that affects Topics Extraction, Lemmatization, POS and Parsing, and Text Classification APIs. Although there are several new features in terms of new functionalities and parameters, the most important aspect of this release lies under the hood and essentially consists of a refactoring of the way in which concept-type topics are internally handled, much more in line with the use of other semantic resources. This lays the foundations for better performance and new features related to the extraction of this type of information. Sty tuned for great improvements in this area in future releases.
The other two great lines of this release are the enrichment of the morphosyntactic analysis with information extraction and sentiment analysis elements (which enable new and richer types of analyses that combine the text’s structure with topics and polarity) and a new predefined classification model.
Here are some details about the developments in the different APIs:
We have just published a new release of MeaningCloud with some new features that will change your way of doing text analytics. As a complement to the most common analytical techniques -which extract information or classify a text according to predefined dictionaries and categories- we have included unsupervised learning techniques that enable to explore a series of documents to discover and extract unexpected insights (subjects, relationships) from them.
In this new release of MeaningCloud we have published a Text Clustering API that allows to discover the implicit structure and the meaningful subjects embedded in the contents of your documents, social conversations, etc. This API takes a set of texts and distributes them in groups (clusters) according to the similarity between the contents of each document. The aim is to include in each cluster documents that are very similar to each other and, at the same time, highly different from the ones included in other clusters.
Clustering is a technology traditionally used in the analysis of structured data. What is so special about our API is that its pipelines are optimized for analyzing unstructured text.
Do you need to understand the impact that social media and news have on your corporate reputation (or your customers’)? Now, a new MeaningCloud API enables you to analyze all that information in real time and structure it according to the dimensions of the most common reputational schemes.
Some customers have been demanding us a precise way of understanding the impact that opinions from social media and other channels have on a company.
Social monitoring tools provide a basic sentiment analysis that in the best of cases identifies that a certain comment (e.g. a tweet) expresses a positive or negative opinion about an entity, and use the aggregated data in diagrams and temporal evolutions. Nevertheless, analyzing a so multifaceted reality as the reputation of a company requires a more granular opinion analysis.
Although it is usually identified with online reputation, corporate reputation is a different concept: it consists of an aggregate of the opinions and perceptions that the public has about a certain company. And it’s a multidimensional characteristic, since those opinions are assessed around a series of more or less standard axes: Financial situation, Product quality, Innovation, Strategy…
So far, corporate reputation has been measured by interviewing and conducting surveys to customers and analysts, and it has been provided in the form of static periodic reports. But the number of customers who are not satisfied with a “snapshot” a posteriori -based on a few opinions- is constantly increasing. Now they wish to add to the picture the impact that social and traditional media have on their reputation, and access that information in a more up-to-date and actionable way to detect potential reputational crises well in advance.
With this MeaningCloud update we have added a new Corporate Reputation API, which enables to include in the reputational analysis the huge amount of spontaneous opinions expressed in all kinds of media (social networks, forums, blogs, news websites) in real time.
This API performs a reputational tagging of text: it receives a document (tweet, piece of news, comment in a survey), detects the mentioned companies, identifies the reputational dimensions involved and extracts the polarity that affects each one of them. All this with a high level of accuracy and granularity, distinguishing opinions about different companies than can coexist in a single sentence. The result is presented in the form of standard tags that can be used to aggregate, relate, detect trends, generate alerts, etc.
Under the hood, the corporate reputation API uses a highly optimized pipeline that incorporates information extraction, automatic classification and polarity analysis techniques. In addition, it is based on standard interfaces and features SDKs that enable to integrate it into any available monitoring application or tool. More information here.
Using this API you can complement the traditional reputational studies with a more agile and immediate analysis of the impact of social media, news, etc., and thus manage more dynamically such an important asset of your company.
And don’t forget that with this MeaningCloud update we include two APIs from our previous product, Textalytics: Linguistic Analysis and Text Proofreading.
If you have recently visited the website of our product Textalytics you may have been surprised by the change of appearance… and name. After commercializing Textalytics for a year and a half and receiving a continuous feedback from our users and customers, we have decided that this new iteration is more “substantial” and help us to bring us closer to the product you are demanding.
During this time, we got thousands of customers around the world using our APIs in business-critical applications. We would like to thank each one of them for trusting us and we assure them that the migration to MeaningCloud will be simple and clear.
There are no major conceptual modifications: we want MeaningCloud to remain the most simple, safe and economical way to convert your unstructured content to actionable information, and so we will keep applying a cloud-based model of APIs. But with this change we want to redouble our commitment to simplicity and personalization, and it results in several improvements.
The most important aspect to emphasize is that all the APIs of Textalytics will remain operational until June 1st, 2015, so at the moment you don’t need to make any changes.
On the other hand, all the functionality of Textalytics you have been using (Text Classification, Topic Extraction, Sentiment Analysis, etc.) has been replicated in MeaningCloud –although with the inevitable change of name in the access points–, so the migration effort is minimal.
But there are also some significant developments: the main one is constituted by the new customization functionalities. The most important thing we have learned with Textalytics is that the key success factor of a semantic analysis application lies in the customization of the tool depending on the customer’s domain (brands, products, departments, competitors…). And, also, that our customers prefer to be autonomous in said customization and to incorporate their dictionaries, models, etc. in a simple way.
That is why MeaningCloud incorporates some new and powerful customization tools based on graphical user interfaces. With this release we make public a tool for creating customized classification models, while tools for the creation of dictionaries and custom sentiment analysis models will be added in the coming weeks. Our intention is to empower users to make them completely self-sufficient, enabling them to customize their analysis without having to learn complicated configuration languages.
Other significant new features, aimed at achieving greater simplicity within the product, are a new user area with a better access to the documentation and the personal settings, and an easier management of consumptions and license keys (which now becomes unique).
Finally, a change in a significant aspect of the service which will be very well received: its price. We have modified the pricing scheme, which now uses the number of requests as billing driver. In this way, we use a unified billing criterion for all APIs. And as a side effect, when it comes to process medium or large-sized documents (i.e. in all cases except Twitter) the charge per request results in a higher volume of processed text. In addition, we extended our Free Plan (which with 40,000 requests per month is the most generous in the market) and introduced an entry step with a lower price, $99. This way, anyone who wants to evaluate or prototype on MeaningCloud, or use it in commercial applications, will not find an impediment in the price.
In response to your requests, we have redesigned our website (both in interaction and information architecture and graphic design) to simplify and optimize the experience of visitors and users.
Besides the mostly informative and open area, there is a section for developers which includes the APIs’ and the integration tools’ documentation, and gives access to a renewed private user area.
We also want to provide you a better experience not only when you access the information about MeaningCloud (its benefits and the solutions that can be developed on it), but also during the subscription process. Now, for the standard subscription modalities, the process is completely automatic and you can pay using PayPal.
Last but not least, we have decided to change the name of the product. We had the impression that the name Textalytics was too associated with a single functionality, text analytics, increasingly more standardized.
Our vocation is to extract meaning from content, i.e. go beyond the recognition of entities or the basic classification –which constitute the standard functionality of text analytics– to analyze sentiment, profile users or detect intentions.
And the possibility of having these capabilities from the cloud and integrate them easily into any application using SDKs and plug-ins –in short, our concept of Meaning as a Service– is such a clear and disruptive value proposition that deserves to be included explicitly in the name. Hence MeaningCloud.
If you are a current user of Textalytics, you’re might be wondering how this change will affect you. We explain it in the following post.
We keep improving the functionality and user experience of MeaningCloud, our horizontal APIs, to offer you a more powerful and easier to integrate text analytics technology.
In this new release, besides incorporating many of the suggestions you have been sending us through our online support, we included several improvements:
In our attempt to make MeaningCloud API the easiest way to use semantics in your application, today we are proud to present our latest development, a MeaningCloud plug-in for GATE.
GATE (General Architecture for Text Engineering) is an open-source workbench for text engineering that makes use of any kind of language processing component, from document crawling to search, and intelligent semantic annotations in particular.
The plug-in provides GATE users a new set of multilingual functionalities, from parsing to entity extraction and sentiment analysis. For MeaningCloud users it would mean an easier and quicker method to prototype full applications including crawling, post-processing or indexing on annotated documents. Besides, if you’re familiar with JAPE rules, it would enable to post-process, mix and match annotations from different processing resources for more complex pipelines. Finally, GATE is ideal for sharing and evaluating pipelines between team members, which increases productivity and produces more accurate results.
Here at MeaningCloud we love Git.
If you have read our posts on sentiment analysis (document-level, feature-level), you’ll have seen that we have started to use gists from Github to share our examples and pieces of code in this blog.
Our aim in MeaningCloud is to make the building of semantics into your applications as easy as possible. Besides our public API, we have developed SDKs to make your life easier. Right now, they are the easiest way to start using our Media Analysis and Semantic Publishing APIs.
Would you like to get your hands really dirty? We have published the code in Github recently!
See: https://github.com/TailorBrands/meaning-cloud
A few weeks ago we were talking on here about how Daedalus (now called MeaningCloud) has explored various API-based business models for marketing its semantic technologies. Our perception was that basic language features are hard to use for many developers —not experts in these technologies— and require them to build solutions through a process of trial and error, in a do-it-yourself fashion, which is slow and inefficient.
Our vision was to offer the market APIs with a more plug-and-play philosophy, which provide a functionality closer to the business, a faster learning curve and, as a result, an increased productivity. That vision has crystallized in our product Textalytics and today we can say that it is now available.
Textalytics (now called “MeaningCloud”) is a SaaS offering that provides a high-level multilingual semantic processing functionality to those developers / integrators who want to develop semantic solutions in an effective, quick and cheap way. Compared with other semantic offerings in service mode, MeaningCloud offers several APIs, each with a specific functionality which is close to an application domain, and SDKs, plug-ins, etc. that make its learning and use much easier, reducing the effort required to obtain results and time-to-market.
