Unlock Digital Signature Licitness for Business Letter in European Union

How to eSign a document: digital signature licitness for Business Letter in European Union

term in the document and this enables different features for example the clustering I showed you before on the extraction of specific relation in this case please consider this is the demo the visual analytics is the demo but just to show you that if I put the the link of this device the system emphasize all the text related to these devices all the sentences sorry link into these devices with the highlighting all the linked elements you can obviously you can click on specific noun so to navigate the text so let's imagine the navigation of the multiple text like millions or billions of texts if you want to analyze in depth the the specific relation the system can propose you also um more in that direction like in this case the subject action object action or different types so actions the movement activities everything having events each sentence has been analyzed an instructor in this tribal types of attributes with all the complements so and establish it is mentioned in the infinity form so all the inflections are analyzed and then manage it in this case complements is when an object or there are different type of complements as you can see here that what other kind of components so in this case you have all the extractions of entities relations in this case you can go in depth on entities so products devices this please remember this is a live demo you can test obviously during the project we can go in depth on the extraction of different uh more specific entity types measures organizations money events also genetic proper noun from the geographical point of view the resection of geographic that is mapped to the database the geonames.org into the types and regarding categories also genetic proper nouns this is a different color means different categories but the categories can be Associated to a different specific taxonomy as you see in this case Finance media topic is a genetic taxonomy we took from iptc.org that is a taxonomy thousands three levels and sometimes four levels as you saw here news agency around the topic is the genetic taxonomy we took from you can see each sentence is related to a specific category that is the column and obviously together with the end user and the partner uh we can train or they can train specific semantic routes to extract more categories in depth so to customize further customize the category hardener for example for different types of art how everything can be applied because they as I mentioned at the beginning of the of the demo the the [Music] the technology is based on a knowledge graph that can be personalized and the knowledge graph in this case an English language is [Music] composed by uh 2 million two million dollar Concepts and six five six millions of different type of relations in this case Bank is a as a different concepts it's sorry different meanings as you can see here red is the verb so to deposit if I click on the first the finance institution that accept deposit and changing you see there is the older hierarchy in the ontology and the credit union is a type of Bank also World Bank everything can be navigated like a this is the least but obviously also in this case we can put more details on and this is the father so generic Concepts and this is the generic so uh United Nation agency is a department that we wrote if I click on below I can see all the type of Bureau below so this enabled the disambiguation algorithm because I showed you as I showed you before the algorithm extracted the meaning not only the keywords so there are different layers as I showed so the ization the keyword and the grammatical logical semantics and morphological so the target is to remove ambiguity so this ambigued and to enhance the Precision in the analysis and one of the steps of the disabilitation algorithm is to navigate the ontology so to extract the the real meaning like a human being said while reading a book or she read the first page perhaps you understand something if read the 20 30 page it can apply reasoning also the first page read previously and this is the the same as the in this algorithm so the target is to remove ambiguity to give more Precision to the end user to the final reader in this case I can show different dimensions the Corpus means all the concepts very often that can be very often found around the department of room so brick work inspect inspection everything so let's imagine a multi-dimensional graph that the algorithm can navigate so to remove ambiguity to enhanced Precision so that the all the tasks after the analysis of the text of multiple text let's imagine multiple text and lies of them can be saved in a decision support system so all this information can be saved in a decision support system so the end user the reader the analyst the partitioners can start reasoning from a multiple amount of structure information coming from the analyzed this is what I like to show you thank you so much and my previous slides you can see also my contacts if you want you can contact me and we can go further uh to show you other live demo specific domain like life science telecommunications cider and so on thank you so much and have a great day