josefrancisco

Thanks to the proliferation of web search engines and their increased efficiency, it has been possible to develop other types of similarity measures based in this type of application.

The main advantage of using search engines is that almost any possible word or meaning can be indexed, so it is not necessary to rely on limited data sources or vocabularies, where the descriptions might be limited or even non-existent.

One of the first works based on Web search engines is the one developed
by Strube. It performs a basic measure such as taking the results obtained when performing a search (hits, page counts) from a search engine and applying the so-called Jaccard coefficient.

Recent research on the use of ontologies in the Semantic Web promises greater interoperability between software agents and Web services, enabling automatic content-based service discovery. However, the heterogeneity in the representation of knowledge represented by ontologies makes interaction between applications that use this knowledge difficult. 

On the other hand, the services produced and described by different developers may use different or perhaps partially overlapping sets of ontologies. Users may store their data in different structures and use different terms to represent the same concept. Therefore, in order to share information, when using heterogeneous vocabularies it must be possible to translate data from one ontology framework to another. The process of finding semantic correspondences between different ontologies is known as ontology alignment and is crucial for the success of the Semantic Web.

The problem of ontology alignment can be described as follows: given two ontologies that describe a set of discrete entities (which can be classes, properties, rules, predicates) it is the process of finding semantic correspondences between them in an automatic way and without needing human intervention, for example, equivalence or belonging.

Ontology alignment a.k.a. ontology matching is a very useful process within the Semantic Web. It can be applied to intelligent agent technology in order to enable them to collect and integrate information from different sources, as well as to search for services on the Web and use them automatically without user intervention. For e-commerce, where ontologies play an important role, information from different and heterogeneous companies can be integrated, unifying descriptions of products, processes, transactions, etc. In addition to all this, the possibility of automating many processes, by providing data with a formal semantics, understandable by machines, is a great contribution. For example, applications from different companies could work automatically with different data sources, even comparing between different ontologies, to transfer or exchange information between them without having to agree beforehand on the meaning of the data.

Ontology alignment techniques are particularly important because the manual creation of correspondences between concepts is not feasible, as it is too time consuming except for very small ontologies. In fact, both alignment and fusion methods allow interoperability between different ontologies. 

However, alignment is much less complex than merging due to the fact that creating and maintaining links between concepts is easier and less expensive than producing a completely new ontology that is consistent with the original ones. Although fully automatic ontology alignment is emerging as the ideal solution for semantic interoperability, the current results provided by automatic methods are still not of sufficient quality. The challenges faced by fully automatic methods are multiple, including differences in vocabulary (e.g. due to synonymy and homonymy), differences in modeling (e.g. due to different models or attribute formats) and different views on modeled reality.

In recent times, we have witnessed a spectacular development of computer solutions based on computer learning. Where the solutions based on computational statistics stand out above all. These solutions are so good that they have been able to improve the state of the art in many domains and fields of knowledge.

However, like all statistical techniques, sooner or later they will encounter obstacles that they will not be able to overcome. Since it is well known that intelligence is not only calculation ability, but also knowledge that can provide facts on which to make calculations. In this sense, we have been working insistently in a survey that covers most of the automatic techniques of creation, exploitation and maintenance of large knowledge bases. Knowledge Bases are really helpful in order to model domain knowledge from experts. These techniques were the precursors of what is known today as Knowledge Graphs.

In the field of manufacturing companies, it is normal to work with processes of transformation and evolution of raw and basic materials. During this process, one passes from obtaining basic materials to totally manufactured products. However, this process is not as simple as it seems. Firstly, because basic materials come in the form of batches. In this way, several batches are mixed with other batches to form an intermediate product. Then other batches evolve over time or are mixed with other batches. Keeping track of this entire manufacturing process is very tedious. And the problems are compounded when there is a problem or defect in a given batch and you have to find out where to go to remove all the products affected by that problem. To model this problem, traditional database management systems based on the relational model are not suitable. Since the nature of this problem is nodes and their corresponding connections For this reason, graph databases are much more appropriate for designing a fully automated tracking solution.

In recent times, we have witnessed an increase in popularity of the smart city and smart village concepts, i.e. the capability to digitalize urban and rural areas using the new technologies such as Big Data and Artificial Intelligence. 

The truth is that the latest advances in such technologies as artificial intelligence, machine learning and big data have made it much easier to develop innovative solutions that can help improve the lives of people living in both urban and rural areas.