L’evoluzione della specie

Business People

– Sensori, telecamere, tre computer installati e uno scanner tridimensionale. Dentro il “corpo” di R1, il primo robot-maggiordomo progettato dall’Istituto italiano di tecnologia di Genova, c’è davvero tutto quello che serve per fare felici milioni di casalinghe stufe di dover pensare ai lavori domestici. Presentato nei mesi scorsi alla stampa, questo umanoide completamente made in Italy è capace persino di portare il caffè a letto e potrà essere in futuro un valido aiuto per sbrigare le faccende di casa. Ma il maggiordomo-robot R1 è anche qualcosa in più: è un simbolo della rivoluzione ormai in atto da anni nel mondo dell’industria globale, dove la robotica e le tecnologie digitali stanno sostituendo progressivamente il lavoro dell’uomo.

Leggi l’articolo



– A margine dell’Aim Investor Day che si è tenuto a Lugano, iFinance ha intervistato Stefano Spaggiari, l’a.d. di Expert System, il gruppo che si occupa di intelligenza artificiale e di ricerca semantica.

Leggi l’articolo di Gianluca Baldini e guarda la video intervista a Stefano Spaggiari

Among the best semantic analysis tools, natural language processing and text mining applications play a key role in today enterprise world. Driven by an increasing need to effectively manage big data, business documents that contain unstructured information as well as social media interactions etc., the market of this kind of semantic analysis tools remains strong (for example, estimates show the NLP market growing from $7.63 billion in 2016 to $16.07 billion by 2021, at a Compound Annual Growth Rate of 16.1%)

A large portion of strategic knowledge is trapped in text in the form of natural language—this means reports, customer call records, clinical trial notes, case law, email and any documents or applications containing human created text content. Being able to access and unlock the business value of this information is a critical challenge that requires tools that can understand and process text in the most effective way.

Semantic analysis tools: some curiosities about Natural Language Processing and Text Mining

In the enterprise data world, natural language processing and text mining have been developing to manage unstructured information (essentially the text formats we listed above.) But while natural language processing processes language to help machines automatically “read text”, the main objective of text mining (or text analytics) is discovering knowledge by transforming unstructured text into structured data. Once analyzed and structured, enterprise data can be shared throughout the organization or with partners for further analysis in the realm of business intelligence projects, customer support activities, social media monitoring, knowledge management requirements, etc.

Cognitive computing for semantic analysis tools

With the growing importance of text-heavy volumes of enterprise data, the business interest in semantic analysis tools  is rising along with the demand for more intelligent technologies such as cognitive computing based on semantic technology. Transforming unstructured text into actionable knowledge requires the capability for reading and understanding language, combined with the power of mining entities, topics, concepts and connections in the most precise and comprehensive way.

Visit our products page to learn how our cognitive solutions integrate natural language processing and text analytics, empowering your ability to use all of your unstructured information for strategic insight and decision making.

Oil & Gas content and unstructured data is growing exponentially – do you have a handle on all your company insights, in-house knowledge and data?

In this insightful and educational webinar, you will learn how semantic intelligence is changing how companies access all of their collected knowledge and how to unlock your unstructured data.

Join us on October 12nd at 3pm ET for this informative and engaging webinar, register here.

Do not miss this unique opportunity to discover how you can turn your unstructured well data into structured information your company can use.

Register here

For a better understanding of what semantics is and how it works, let’s look at some simple examples of semantics in the context of how the human brain works.
The average human brain has a general knowledge of the world, and it uses this knowledge, as well as that of past previous experience to understand words and the relationships between words and sentences; semantic technology can do the same.

While we know a lot about how the brain operates, even centuries of study and research have not been able to fully unravel all of its functions, such as how it stores and retrieves memory. As we’ll see in the examples of semantics below, semantic technology takes inspiration from the way the human brain processes information to understand a text.

Inspired by this infographic that maps how different areas of the brain handle different tasks, we created a simple and unscientific parallel representation of how semantics approaches the processing of a text.

Examples of semantics: How semantics processes a text

●    External stimuli: Information sources—text documents, web pages, social media and emails, etc.—while potentially diverse in terms of content and context, are nonetheless information that must be ‘processed’ to be understood.
●    Neurons: These are the pieces that make up the semantic algorithm, which allows information to pass through the various stages of linguistic analysis. It is through this advanced algorithm that semantic technology is able to understand language in the same way that people do.
●    Hippocampus: Extracting and storing concepts requires determining the semantic context for the proper disambiguation of terms.
●    Just as our brains pull from previously read or memorized information when we read, semantic technology is able to accurately determine the context of information, and therefore, interpret it with the precise meaning.
●    Cerebellum: In semantics, this is the semantic network, a conceptual map made up of words and all of their different meanings and connections to other words. This constitutes the “seat” of learning, knowledge and all the functions involved in language comprehension.
●    Amygdala: Just as this area identifies emotions and feelings, semantics takes cues from language and context to understand when a text conveys feelings of fear or happiness, sadness or satisfaction.

With these modules and examples of semantics we can understand why semantic technology is the most advanced approach to language processing, making many practical applications possible, from search engines to natural language interfaces, the extraction of specific data to the categorization of content.  For those more complex tasks, we’ll still need to use our brains!


– Historically, technology tools have offered solutions that provide conclusions based on the analysis of static data and transactional data associated with specific banking activities. The different types of data vary greatly. To name a few there is big data, data from diverse legacy systems, shared file data, disparate data files from incompatible systems, incomplete data, qualitative data, numeric data, unstructured data, data using competing terminology to describe the same activities, and more. To meet compliance requirements, financial institutions, their financial crimes and compliance leaders and their respective teams must have tools to make sense of it all. To meet the high bar of compliance, financial crimes units have often relied on increasingly larger teams of people to manually comb through reams of information in an effort to address the challenges with “tribal wisdom,” professional experience, human reasoning and brute force.

Read the article


Expert System is pleased to announce that it will take part in the 3rd annual Insurance Analytics Europe conference held at The Grange Tower Bridge, London, October 5-6.

Insurance Analytics Europe

As the largest insurance data and analytics event in Europe, Insurance Analytics Europe brings together more than 250 executives in business and tech.

Nicky Singh, Expert System’s VP Insurance Innovation, will present a joint session on how cognitive computing will disrupt and innovate the insurance sector.

Our cognitive computing technology is production-ready today and  offers an immediate opportunity for insurance providers to deliver  business value with limited resources.

Visit us in booth #11 for a live demonstration and to learn more.

For more information about Insurance Analytics Europe, visit the website

L’accordo, stipulato con l’azienda specializzata in AML (Anti Money Laundering) NominoData, prevede l’impiego dell’intelligenza artificiale di Cogito nell’analisi dei big data e nella text analytics per migliorare la gestione del rischio

– Expert System, società leader nello sviluppo di software semantici per la gestione strategica delle informazioni e dei big data, quotata sul mercato AIM di Borsa Italiana, ha stipulato una partnership strategica con NominoData LLC, società statunitense dedicata all’offerta di soluzioni per la gestione dei problemi aziendali legati a identità e reputazione, rischio e compliance.

L’accordo prevede la combinazione della profonda esperienza di NominoData nella copertura dei rischi (dalla gestione dei rischi operativi alla prevenzione e all’indentificazione delle frodi, all’anti money laundering) con le soluzioni di intelligenza artificiale di Expert System ideate per supportare le organizzazioni a migliorare i processi decisionali. L’arricchimento dell’offerta di Expert System con i dati messi a disposizione da NominoData consentirà ad entrambe le aziende di rafforzare il posizionamento non solo nel mercato creditizio e assicurativo ma anche al di fuori dell’ambito finanziario.

Robert A. Goldfinger, Presidente di NominoData LLC, ha così commentato: “L’integrazione della piattaforma di Expert System con i dati messi a disposizione da NominoData rafforzerà la capacità dei nostri clienti di mitigare il rischio attraverso l’intelligence dalle fonti aperte. Nell’ambito della gestione del rischio, infatti, l’identificazione di informazioni rilevanti su individui ed entità può consentire ad analisti e investigatori non solo di soddisfare i requisiti normativi, ma di sviluppare anche opportunità di avanzamenti e miglioramenti operativi.”

L’offerta delle due aziende farà leva sull’impiego di un esteso set di dati, tra cui: notizie negative, rilevanti per la gestione del rischio, provenienti da fonti aperte (stampa, fonti governative, blog, ecc.); database con informazioni su persone esposte in ambito politico (PEP data, Politically Exposed Person – data); disposizioni governative e watch list provenienti non solo dagli USA ma da tutto il mondo; informazioni sul business dei 24 stati americani in cui è  stato legalizzato il mercato della marijuana; tabelle con cap, nomi di città e paesi in aree geografiche considerate ad alta densità criminale o legate al traffico di droga (HIFCA, High Intensity Financial Crimes Area HIDTA data; High Intensity Drug Traffic Area); un ampio insieme di varianti di nomi; e i file dei Panama Papers con oltre 150K di informazioni su persone listate nei documenti.

“Siamo soddisfatti della nostra partnership con NominoData”, ha dichiarato Bryan Bell, EVP, Market Development della sussidiaria americana del Gruppo Expert System dedicata al settore delle aziende private. “Unendo le potenzialità del cognitive computing di Expert System nell’analisi dei big data con il ricco patrimonio di dati di NominoData, potremo offrire soluzioni innovative e molto produttive a supporto della mitigazione dei rischi.”


Expert System e NominoData presenteranno l’offerta congiunta nel corso del più importante evento mondiale dedicato ad antiriciclaggio e criminalità finanziaria (ACAMS 15th Annual AML & Financial Crime Conference, Las Vegas, 26 – 28 settembre 2016).

Regardless of industry, the overload of information facing most organizations today is a drain on both individuals and the enterprise itself. When it comes to separating the useful information from the irrelevant, document classification is a worthwhile tool that can reduce the cost and time of searching and retrieving the information that matters.

How does document classification work?

Document classification is an age-old problem in information retrieval, and it plays an important role in a variety of applications for effectively managing text and large volumes of unstructured information. Automatic document classification can be defined as content-based assignment of one or more predefined categories (topics) to documents. This makes it easier to find the relevant information at the right time and for filtering and routing documents directly to users.

Document classification has two different methods: manual and automatic classification. In manual document classification, users interpret the meaning of text, identify the relationships between concepts and categorize documents. While this gives users more control over classification, manual classification is both expensive and time consuming.

Automatic document classification applies machine learning or other technologies to automatically classify documents; this results in faster, scalable and more objective classification. There are at least 3 approaches:

By classifying text, we are aiming to assign one or more classes or categories to a document, making it easier to manage and sort. This is especially useful for publishers, financial institutions, insurance companies or any industry that deals with large amounts of content. An automatic document classification tool can realize a significant reduction in manual entry costs and improve the speed and turnaround time for document processing.

Why Semantic Intelligence is the best option for document classification?

Semantic technology processes and interprets content by relying on a variety of linguistic techniques including text mining, entity extraction, concept analysis, natural language processing, categorization and sentiment analysis. Semantic technology allows the automatic comprehension of words and entire documents, and understands the meanings of words in context.

As opposed to keyword and statistical technologies that process content as data, semantic technology is based on not just data, but the relationships between the data. This ability to understand words in context is what makes automatic classification possible, and enables not only the management of large volumes of data, but the ability to optimize it for even further analysis and intelligence.

Expert System, società leader nello sviluppo di software semantici per la gestione strategica delle informazioni e dei big data, quotata sul mercato AIM Italia, organizzato e gestito da Borsa Italiana, comunica ai sensi dell’art 25 del Regolamento AIM Italia, la nuova composizione del capitale sociale risultante a seguito dell’integrale sottoscrizione delle massime n. 2.609.552 azioni ordinarie di nuova emissione offerte nell’ambito dell’aumento di capitale deliberato  dall’Assemblea straordinaria del 28 giugno 2016 e conclusosi lo scorso 13 settembre 2016.

Si riporta di seguito la nuova composizione del capitale sociale (interamente sottoscritto e versato) a seguito dell’attestazione di avvenuta variazione depositata e iscritta presso il competente Registro delle Imprese di Trento:


  Capitale sociale attuale Capitale sociale precedente
  Euro Azioni Var. Nominale Euro Azioni Var. Nominale
Totale di cui 276.703,30 27.670.330 272.928,64 27.292.864
Azioni ordinarie 276.703,30 27.670.330 272.928,64 27.292.864


Nell’ambito dell’aumento di capitale sono stati altresì emessi complessivi n. 2.497.552 “Warrant Expert System S.P.A. 2016-2018”. Si riporta, quindi, la tabella riepilogativa del numero dei warrant attualmente in circolazione:

  n. Warrant in circolazione
Warrant Expert System S.P.A. 2016-2018 2.497.552


A seguito delle sottoscrizioni dell’aumento di capitale, la Società comunica che, dalle ultime risultanze in suo possesso, il proprio azionariato risulta composto come segue:


Nominativo N. azioni Expert System Partecipazione (%)
Paolo Lombardi 2.781.886 10,05%
Stefano Spaggiari 2.758.885 9,97%
Marco Varone 2.736.886 9,89%
Luxid Sarl 1.374.761 4,97%
Mercato 18.017.912 65,12%
Totale 27.670.330 100,00%


L’emittente provvederà ad aggiornare la composizione dell’azionariato ed a darne tempestiva informativa qualora dovessero pervenire comunicazione dagli azionisti.