CASE STUDIES

This project is aimed at ferrying customers from traditional to digital sales channels.

The solution allows you to direct the efforts of the sales team towards those customers who may be more sensitive to this option by identifying them with an artificial intelligence solution.

The proposed solution must be as independent of arbitrary parameters and thresholds as possible.

The implemented solution analyzed, within the customer’s datalake, the information that could describe the customer base in terms of absolute and digital purchasing habits.

From this dataset, standardized and reduced in dimensionality by PCA, outliers have been eliminated using the DBSCAN algorithm. On the remaining data, clustering algorithms were developed, aimed at finding regularity between customers: in particular, given the careful choice of the dataset, one of the two clusters into which the population was divided was formed only by customers who already have a tendency to operate on online channels. The centroid of this cluster was taken as “model online customer” where the distance from the model was calculated for each point.

The result makes it possible to identify clusters within the customer base with a greater propensity to operate on online shopping platforms.

Semantic Analysis, based on Artificial Intelligence techniques, allows the machine to understand texts and extract information.

Our Semantic Search Engine allows you to scan large amounts of documents to discover information and answer specific research questions. Our solution identifies facts and relationships that would otherwise remain hidden in the mass of textual big data. Once extracted, the information is converted into a structured form that can be represented through intuitive graphic objects.

In detail, our Semantic Search Engine allows you to:

• extract text, tables, images and other elements from newspaper articles, scientific publications, manuals, data sheets, reports, contracts, etc;
• translate, by OCR, images containing text into actual text;
• submit the texts obtained to semantic analysis, with the dual purpose of obtaining the indexing of the contents and of reconstructing the text in a web friendly form;
• create a question answering system that automatically answers the questions asked through natural language, extracting the contents from the previously created knowledge base.

Answering questions is not anymore looking for a string in a text, but for a concept in a piece of knowledge (ontology), according to the context.

Our solution is built for the following sectors:

• Publishing
• Accounting firms, notaries, lawyers
• Engineering studies
• Scientific organizations

Every day, the world’s financial organizations has to face the threat of fraud. Our line of Artificial Intelligence solutions includes machine learning tools to identify fraudulent transactions in the financial markets. Our system is able to identify:

• Whether a trading day for a stock is normal or not, based on its past history
• If a customer has performed anomalies in his movements, with respect to its past history

These problems are attributable to the class of so-called anomaly detection problems, that is, related to the discovery of anomalous situations submerged in a background of data considered normal.

To obtain a good accuracy of the algorithm it is necessary to have a dataset rich in information on the past history of securities and customers. For the titles, all the information relating to the individual market days is required, such as average price, volatility and quantity traded. While for customers it is necessary to have all possible information relating to personal data, demographics, portfolio data and the history of movements.

The days of stock trading or the movements of a client are thought as points in motion in an appropriate space. As the days pass, the point will draw a trajectory in this space, which will mostly remain confined to an area, occasionally exiting it in conjunction with events that disturb this motion. The zone in which the point spends most of its time represents the normality of the conditions described by the coordinates, while the occasional points outside this zone represent anomalous situations.

Unsupervised networks are trained to search for points of anomaly regardless of parameters and thresholds, only by referring to the past history of the stock or customer.

Our solutions are currently in use by major Italian and international financial realities.

The provisions of B.d.I. of 5/12/2018 on governance and control mechanisms for retail banking products introduce the protection of consumers of banking services, through a set of rules governing the life cycle of the product.

There are no specific indications or prescriptions at the regulatory level regarding the criteria to be used to define the reference market, positive and negative. During the entire life cycle of the product, internal procedures must ensure that intermediaries take into due consideration the interests, objectives and characteristics of customers, the typical risks of the products that can cause prejudice for customers, and possible conflicts of interest.

Therefore, the identification of the positive and negative target market remains essential, the classes of customers for which a product can be considered suitable or unsuitable.

Our solution allows the preliminary identification of the set of data wich describe the customer and the products, describing on the one hand risk propensity and degree of financial literacy and on the other the degree of risk and complexity.

The obtained datasets are submitted to clustering, an unsupervised method that allows the dataset to be separated into homogeneous subsets.

In this way the initial pairs (Customer, Product) are remapped into families (Customer Cluster, Product Cluster) that define the target markets. Finally, the target markets are submitted to machine learning methods that identify any outliers, discovering the numerical consistency of off-target customers.

It should be noted that this approach can also be used in a proactive key, referring a new customer to a cluster of customers and proposing the products related to the most appropriate cluster of products.

The financial services industry is undergoing a significant transformation in the way financial management and advisory services are requested and delivered. This evolution is due to a variety of factors: new regulatory parameters, customer demographic changes and, above all, technological advances. These changes are seen at a time when the need and demand for financial product advice has never been higher.

The advice provided through a robo-advisor is customized on a profiling performed on the customer. This profiling serves to define the characteristics of the investor, mainly in terms of risk propensity or aversion,

on financial objectives and / or constraints. Once the customer has been “framed”, an allocation of investments that is appropriate to his profile is proposed to him.

Our solution can preliminary identify the set of data that describe the customer and the products.

Due to the capabilities of Artificial Intelligence, it is possible to build even very complex and high-dimensional representative models.

Algorithms are built to group financial instruments on one side and customers on the other, and then cross the groupings and determine the investment. Neural networks associate model portfolios with customers.

This tool acts as a valid support in developing investment strategies and in ensuring compliance of the latter with legislation.

Development of a Lead assessment tool to support the sales force.

This solution uses artificial intelligence algorithms to assign a probability of winning for each lead / negotiation, in order to focus the sales force on winning contacts, and at the same time look for preventive strategies to increase the chances of success.

The strategy proved to be successful because it allows direct use of significant but not quantitative data, therefore impossible to insert into a mathematical model.

This solution involves the adoption of an artificial intelligence tool to “read” the answers of a 350-question test developed by psychologists specialized in recruiting.

The goal is to use artificial intelligence to discover relationships nested in large amounts of data to provide evaluators with an objective feedback tool.

In this project we have developed a deep learning-based machine vision system that interprets images in the X-ray band of food packaging.

The contribution of artificial intelligence is fundamental, allowing us to build a simple, reliable and flexible system, characterized by rapid reconfiguration and machine learning.

This project aims to optimize sales of consumables that are repetitive in nature.

The goal is to use the ability of predictive analytics to discover patterns in customer sales to optimize the time the sales force contacts customers and predict the need for consumer products.

The result is a better lead prediction and sales optimization.

This solution is dedicated to public administrations, public security forces and companies that need to inspect videos of vandalism, events, accidents, thefts, etc.

The implemented system is able to analyze videos and identify people, vehicles or events, such as fire, in order to support the police in the control of the territory.

The system is able to make detailed analyzes, for example by searching for the face of a human figure, when this is detected, or by searching for the license plate of a vehicle.

This solution allows to determine the probability of abandon of a customer through Artificial Intelligence techniques. An accurate prediction of the probability of a customer’s abandon (churn prediction) has an impact on various aspects of the business, including: customer-oriented proactive marketing, sales forecasting, pricing models oriented to customer retention.

The solution can predict whether or not a customer will leave by considering this as a classification problem and building a descriptive dataset of customer behavior from the sources in the company. The answer is given in terms of the customer’s class (churn / no churn) and associated probability.

It is also possible to give an estimate of which variables in the dataset have greater weight in one or the other class, obtaining information on where it is most advantageous to act to prevent abandon.

The solution provides a system that collects an order from a restaurant table via mobile device, validates it and routes it to the kitchen and the cashier.

The solution grants:

1. Voice recording

2. Speech to text conversion via artificial intelligence layer

3. Text validation via NLP

4. Packaging of the message to the management system