Version 1.12
Software derived from the Urban Soil Analyzer, by the same author, registered in Instituto Nacional de Propriedade Industrial (INPI), as published in Revista de Propriedade Industrial (RPI) 2363, process BR 51 2016 000239-5.
It’s a software that uses artificial intelligence based on Artificial Neural Networks, to create an effective decision support system, that can be applied thoroughly to every branch of human existence.
The Analyzer works like the human brain identifying similarities between individuals. It uses statistical and complex math tools to rank them according to their similarity or affinity. It has the excellent ability to learn, that is, to acquire knowledge over time.
In many areas of knowledge, because it has a huge ability to rank objects, people, places, events, behaviors at all, processes, theses and much more.
This ability allows the Analyzer to sensibly collaborate in important society areas, as in the education, health, food, work, living, transportation, recreation, safety, social security, assistance to the disadvantaged, arrecadation and tax inspection, debt collection, perception of consumption profiles, support to sale strategies, priorization and grouping of administrative and judicial processes, among others.
The huge literary universe needs an optimized model of similarity that make it possible to identify the profile of each reader and to infer what kind of reading he or she would like to read or to know.
Once the publications for sale on a website have been registered, for example, the surfing behavior of the user will be observed registering the pages of interest and the clicks made. From there, the neural technology would be able to identify behaviors, and infer about reading habits that lead to the indication of the probable desires of each reader, offering him the indication of the best books for people with the checked profile, potentializing the sales.
On the other hand, the registration of customers and their respective purchases made in the physical or virtual store, would enable the Neural Analyzer to search among the large range of books, periodicals and other publications available, those ones that may interest the customer, again increasing sales and the profit of publishers or bookstores.
Similar procedure can be used in public or private libraries for the identification of similar books that could interest each reader or student. In this way, the Neural Analyzer may reduce the effort of search for matter linked to the researched subject, bringing also a great ease in the identification of diverse sources of research.
As above mentioned the use of the artificial intelligence mechanisms in the Neural Analyzer gather many of the areas related to reading as well as to its businesses.
To leverage tax collection, the Neural Analyzer can be used to direct fiscal actions of audit and intelligence based on the classification of taxpayers who are most likely to pay taxes.
After collected various data on noncompliance of the principal and accessory tax obligations, contestations in administrative appeals, results in administrative or judicial charges, assiduity in assumed installments, as well as cross-information with external sources that indicate the existence of wealth (such as real estate registration, transit departments and others), make possible to infer, with a reasonable degree of certainty, the group of taxpayers with the best perspective of collection.
Similarly, taxation can also be surgically directed to those taxpayers where there is a higher rate of tax return, that is, of effective collection resulting from tax assessments.
This topic was separated from HEALTH, due to the great importance and scope of the topic today.
Occurring events in which there is an uncontrolled spread of illnesses, as happened in December 2019 with the New Corona Virus, the use of Neural Analyzer intelligence can provide expressive support for decisions aimed at reducing negative consequences, through one or more of the following actions (in alphabetical order):
ALLOCATION OF BEDS
DIAGNOSIS OR PROGNOSIS
MANAGEMENT OF MEDICAL, HUMAN AND OTHER RESOURCES
INDICATION FOR THE TRANSFER OF PATIENTS BETWEEN HOSPITAL UNITS
INDICATION FOR ADMINISTRATION OF MEDICINES AND THEIR ASSOCIATIONS CASE BY CASE
PREDICTION FOR THE EVOLUTION OF THE PATIENTS' CLINICAL SITUATION
HOSPITAL DISCHARGE PREDICTION
CONTAMINATION ROUTE PREDICTION (BETWEEN NEIGHBORHOODS, CITIES AND OTHERS)
PATIENT SELECTION FOR PROPER DISTRIBUTION AMONG HOSPITAL UNITS
Just to exemplify the use, let us take the application of the technology for one of the examples above, the analysis of possible new cases of the disease, or the prognosis for a certain patient.
Without using technical-scientific preciosity, let's use as an example a modeling built with the following data, or part of them:
PATIENT'S PERSONAL DATA
Age / Age group / City / Education / Gender / Marital status / Neighborhood of residence / Neighborhood of work / Patient died? / Professional activity / Social class
HEALTH FRAMEWORK
Cardiac? / Chronic kidney failure? / Diabetic? / Patient exercises regularly? / Any breathing problems? / Ex-athlete? / Hypertensive? / Immunodeficient? / Recently transplanted? / Smoker?
DATA ON THE ENVIRONMENT
Number of infecteds in the family / Number of people in the family / Occured conscious contact with an infected patient?
MOMENT DATA
Date of disease detection / Length of hospitalization (in days)
SYMPTOMATIC DATA
Breathing difficulty (none, mild, moderate or severe) / Continuous days of fever / Cough intensity (none, mild, moderate or severe)
LABORATORY EXAMS
Covid-19 test result / IgG positive? / IgM positive? / RT-PCR positive?
The inferences generated from the analysis of the collected data must be constructed based on the orientation of the business area, which leads to the achievement of more objective conclusions and directed to what is to be analyzed.
In the specific case above, the indication is obtained not only for the prognosis, but even indications about the need for hospitalization, with the respective probable time of hospitalization.
In addition, as the Neural Analyzer has its own tool for generating forms for data collection with tablets, cell phones and even computers, data feeding is greatly facilitated.
For more information, contact us by the e-mail contato@analisadorneural.com.br or through our social networks.
We want to help!
Charging, and especially receiving, requires knowledge of the debtor and the debt, of its personal characteristics and the situations involving the payment itself. The Neural Analyzer is an excellent tool for analysis and for understanding the collection and the receipt, including all the elements that form the complexity of the situation.
It’s fundamental to the Analyzer to know the debt, its original value, its amount with interest, fine and monetary atualization included, its age, the benefits offered to its payment, the installment options, as well as the debtor himself and all the characteristics that qualify him, such as gender, age, schooling, monthly income, recurrence factor, the situation in existing credit protection companies, and others, so the Analyzer can identify the behavior patterns and provide support elements that help to decide how to charge it. That’s what it does the best!
The Neural Analyzer finds one of its most notable fields of use on education, providing great results on allocation of human and materials resources, both in the private and public areas.
The Analyzer is responsible for huge gains in productivity and efficiency, both in the daily management and operational actions, improving the selection of faculty in relation to the school profile, the way the classes are configured, aiming the formation of similar classes, besides improving the definition of good food for a proper nutrition, the conflict resolution and prevention between classes or individuals, the preparation of tests, the selection of literary texts, the suggestion of books to read according to the student profile and the library collection, also helping in many other aspects of school life.
On the large area of finances, the field of application of the Neural Analyzer is very wide.
In the credit analysis, for example, the extensive cataloging of clients and the results obtained with loans, allows inferring the possible success or failure of the delivery of resources to a particular potential customer.
Signing features such as the company's opening date, the relationships between short-term and long-term indebtedness with gross annual sales, the number of members and subsidiaries, the existence of restrictions in credit protection companies, the total of movable or immovable property, and other business features will help the Neural Analyzer to provide safe indicators that subsidize the granting of customer credit, intending to protect the financial institution from disastrous decisions and operations that could lead to undesirable losses.
In managing applications, inferring the most appropriate time, potential return, risk and other elements related to a particular stock market application, for example, based on the success or otherwise of previous applications, considering their characteristics and occasions, represent a precise field of use of the Neural Analyzer, which can provide you and your business with the most consistent decision support mechanisms.
Also in decision support systems on financial and economic investments the Neural Analyzer in general tends to be of great value, especially given the huge range of possibilities and variables involved when it comes to this subject.
The remarkable mathematical analytical capacity on which its algorithm is based gives a special ability to identify the best investment option, considering the variables set and recorded in relation to previous events.
Whether it is for stock market or real estate investment analysis, for example, the software has a great prospect of aid by identifying similarities that it has the ability to address.
Contact us. We will be pleased to serve you!
Analyze the profile of your future investments faster and more assertively!
The Neural Analyzer can be used in many sectors linked to the food branch, as in supermarkets, restaurants, snack bars, public house, distributors of food and beverage, among others.
In supermarkets, for example, after the payment, the shopping made by the client can provide a database that allows to infer which pairs of products are usually acquired together, in order to approach the shelves that contains them, aiming to increase the probability of consumption.
In restaurants, the habits of consumers and their families can be classified. Thereby, the manager will make decisions based on the maximization of the economy with purchases, promotions and many awards to his employees from the identification of peculiar habits. The behavior of consumption can be observed at determined days or schedules, or related to different groups (singles, couples without children, couples with children, young friends, adult friends, seniors etc)
In food distributors, wholesale or even retail sales, once ranked by similarity, the manager can improve dealings with suppliers, reduce operating costs, avoid shortages of goods, and leverage profits.
Thus, as you can see, the use of artificial intelligence is also very embracing and versatile in the food market.
In health the Neural Analyzer finds a vast field of application as a system to support medical or administrative decisions, as exemplified briefly below:
After the patient has informed his/her physician about a set of clinical symptoms of the disease that affects him, also with possible anamnesis data, the complex mathematical mechanism processed by the Analyzer in relation to previous diagnoses will help the physician to find out the indication of the possible disease. Depending on the data types available, the physician may obtain even the suggested treatment inferred from the comparison with similar success cases.
As the complexity and evolution of the disease make the treatment of the patient more and more critical, the agile and correct prescription of drugs becomes determinant for the recovery and maintenance of the patient's life.
Focusing in the patient recovery evolution, the Neural Analyzer can be used by physicians to identify which drug or set of medications has achieved the highest success rate in previous similar treatments, as well as which procedures, in theory, should be observed to avoid complications in the clinical situation of the patient.
At the reception in health care units or in hospitals, the risk classification that tends to indicate the urgency needed for the patient’s care can be optimized significantly by the use of neural classification technology that, based on the analysis of similarity with previous cases, allows to infer about the real gravity of the case of the patient.
The erroneous sending for a medical professional with a different specialty from the one that should be properly indicated for the consultation of the sick patient, besides delaying the attendance and generating rework by the forwarding to the appropriate physician, can compromise the need for urgency and medical measures to the patient.
In this sense, as in the risk classification mentioned before and based on the success of previous referrals, the Neural Analyzer allows to indicate the appropriate medical specialty to treat patients with the reported characteristics.
On MANUFACTURING, a huge industry challenge is to prevent failures in machinery and equipment. In this respect, it is fundamental to use the Neural Analyzer to detect the probable occurrence of defects before they occur, allowing early prevention actions, such as timely replacement of parts that, despite its low costs, when defective, can lead to millions of losses to the operation and sales of manufactured products.
With this clustering technology, the proper classification of machinery, its sensitive parts, its operating and usage rules, the maintenance criteria, the review periods and other characteristics, including dates and circumstances in which problems or defects occurred before its operation, would make possible the prediction of accidents, based on the need for preventive corrections in order to avoid interruption of production.
Similarly, in the ASSEMBLY of parts and pieces, the use of the Neural Analyzer can help to identify procedures that need revision and improvements from similar defects observed in the final product, supporting the decisions necessary for the correct allocation of labor, work shifts, setting, auxiliary equipment and others.
As seen above there is a large usage of the artificial intelligence technology to prevent failures in the industry, as well as to antecipate assembly problems in many ways.
By observing and registering the conditions involved in manufacturing, such as temperature, pressure and others, as well as observing the results achieved as the final product produced, it is possible to identify the best relationships between such conditions, which lead to a higher yield in production and, consequently, to a desired reduction of waste.
As in activities for which there is a wide range of possibilities to consider, in inspections in general, the use of the Neural Analyzer brings uncountable benefits.
From the identification of similarities between previous inspections and their results, the Neural Analyzer can infer with considerable degree of certainty, the potential of new inspections results, supporting decisions regarding to who to inspect, when, which resources to use (human, material, logistical etc.) as well as the possible outcome to be achieved.
Looking towards against tax evasion of public resources arising from the omission of revenues and the correct payment to the government, related to industrial, commercial or service transactions, including financial ones, is an area of great public sector study, dedication and investments.
Through complex mathematical mechanisms and careful analysis of data related to previous tax inspections, the Neural Analyzer infers the results of new inspections even before they happen, thus avoiding unnecessary allocation of human and material resources, among others, aimed at conducting activities with little or no fiscal result.
Otherwise, the correct targeting of actions, enhances the results and maximizes the revenues from inspections, whether taxes, fees, contributions or fines in general.
Seek to know. Invest in technology to raise money. Charge better, in a cheaper and faster way!
The client's proper defense in judiciary cases emerges for the operators of law and regulations as an immeasurable challenge, for which only victory is desired.
The structuring of legal workpapers usually involves knowledge and mastery of an enormous amount of laws, decrees, opinions, judgments and the whole jurisprudence itself, leading to a very difficult choice to select the perfect thesis to defend the party.
In such a vast universe, the use of the Neural Analyzer and its meticulous ability to classify situations by similarity can bring an expressive shortening of the way to find the fundamentals that lead to a noticeable chance of success in obtaining the demand.
Having registered previous situations with their peculiar characteristics, the neural technology infers the best defense, accusation or decision models to be used in the administrative or judicial petition to be elaborated. It will be enough, so far, to inform the characteristics of the situation to analyze, in order to establish the connection, determining the similarity level between the current situation and the others.
To company use, it is generally challenging to define which products to launch and how to do it, keeping in mind the best possible acceptance and satisfaction to its target audience.
Determining and measuring the orders to be made to suppliers is another point that, if well planned, tends to maximize profit margins.
The Neural Analyzer is an excellent tool to, based on the success of product already launched and orders already made for suppliers, assist managers to take better advantages in their decisions, reducing financial, time and labor costs.
1) On resale, it can be used to:
- Define the consumption profile of the existing customer portfolio, providing a better targeting of the products that should be offered for each specific profile in order to maximize the sales.
- Identify new potential customers, by similarity with existing customers, for specific product or product line.
2) In consumption, it can be used to:
- Understand how products circulate in the network. When a new micro-franchisee is sponsored its possible to infer which type of product he/she will use more and how often.
3) In the consumer market expansion, it can be used to:
- Identify prospects with the highest potential to become new micro-franchisee, that is, to be recruited. It's also possible to infer the prospects more likely to reach high levels of leadership, to reach minimum monthly consumption, among others.
- Estimate the chances of micro-franchisee going to the next level/graduation/pin, based on several important variables as number of people in his/her team and their levels/graduations, number of people buying the minimum consumption in the network, moment of the cycle when the analysis is done, etc.
Frequently it can be seen public servants surrounded by endless piles of paper. They are administrative processes, memos, diligences and many other kinds of papers looking for some action.
The detailed analysis in each situation shows itself to be slow, and sometimes complex and boring, especially due to the wide range of control variables, such as the subject matter of the request, the facts on which it is based, the procedural deadline, the origin office or sector from the appeal and others.
An expressive problem that retards the solution of these lawsuits is the great human inability to group together complex problems, giving them similar solutions, applying analogue understanding and measures to documents and situations that looks like one to the other, being sometimes even identicals.
At this sight, the Neural Analyzer has a huge mathematical ability to group such problems by similarity, tending to advance and accelerate the analysis and resolution of these demands.
Whether in the sphere of the Executive, Judicial or Legislative authorities, the grouping using the neural technology deserves to be known. More than that, deserves to be added without delays, in the internal processes of the public service, aiming to the immediate gain of efficiency in all the related jobs.
The police investigation finds in the Neural Analyzer an excellent tool to identify, with expressive degree of certainty, the authorship of crimes and offenses.
From the analysis of preexisting data that indicate such subversions, the neural technology allows to infer by likeness the probable author of the malfeasure, facilitating and accelerating the investigation in their respective collections of evidence, testimony, searches, seizures and others, before it is actually possible to conclude on who executed the punishable act.
Take, for example, the use in flagrant offices, where the complaint of a committed crime, such as a theft, robbery or murder, is accepted. From the prompt collection of information about the facts of the event, the police authority, by sending these characteristics to the Neural Analyzer, depending on previously recorded data on similar facts, may obtain an accurate indication of possible authorship, in order to immediately arrest or apprehend the perpetrator of the crime in question.
Sales at all involve some elements that require, by themselves, complex control and administration: the product itself (type, brand, specification, unit, etc.), customers, sellers, the time or sale moment, the point of sales characteristics, the collection of taxes, exchanges or returns, delivery logistics, billing and collection, risks and marketing, among others. All these data set up in an adequate model result in learning that give the businessman an intelligent vision about the best decision to make.
Identifying or maximizing the profitability on each sale is a task that needs a high ability, knowledge, study, historical records and, sometimes, taking risks on which there are little or no control.
The Neural Analyzer, as a tool of a high mathematical power and capable of generating intelligence through learning with historical data analysis, can infer several elements about the sale and the factors related to it. It's artificial intelligence.
Low complexity inferences that can be obtained by means of sales data modeling, could be for example:
1. What seller’s profile is most appropriate for saling a certain product;
2. What kind of sale’s actions (advertising or other) are best suited for the sale of a particular product on a specific date or period?
3. What are the most efficient after sales mechanisms for getting customer’s loyalty, and in what situations?
4. To what products and customers the delivery prove to be effectively relevant, and in what situations?
The greater the number of variables involved in the analysis, the greater the complexity found to establish a mathematical and logical correlation between them. So, higher complexity inferences could be:
1. Which product groups sell better when offered together at a specific moment (holiday eve, fortnight of the month, etc)?
2. Who are the sellers that get the best customer’s loyalty, with what type of product and moment, using which sales strategy?
3. What kind of charge may get the better achievement? The one on what customer profile, at what time, through which way and offering how many advantages for the payment?
4. In what period of the day, week or month happens the sales with the greatest return or change request made? By sellers with which profile or using what point of sales, and in what type of product?
In addition to the innumerable possibilities of using our technology in actions directly focused in sales, it also has large application in the negotiation with suppliers, allowing even to indicate, based on data from previous negotiations, which behavior or approach is most appropriate to be used in each new negotiation.
Much may be also obtained in relation to identifying the most suited point of sales for offering a particular product, even directing the purchase of related products for which the buyer’s profile or the moment of purchase indicates a greater chance of the aggregated sale to be accomplished.
Certainly, beyond the applications suggested above, your business can glimpse many others from his daily knowledge and practice, simply by appropriately modeling the data, generating the desired intelligence and achieving greater gains to the enterprise, reducing losses and decreasing risks of all kinds.
By analyzing the characteristics of travelers, their families and their businesses, as well as the places they travel to, the accommodations they pay for, the possible rent of vehicles, the length of stay in each place, the amounts invested in leisure or business as well as many other elements that compose the trip, our system generates knowledge and, with it, intelligence capable of identifying profiles of people or groups, and with this, infers with a incredible degree of certainty, those services that should be offered to a new client, or even to an old one already habituated to contract packages with the company.
The complexity generated by the wide range of combinations that define the traveler's profile, together with the history of their travel options and the peripheral services added to them, finds in our product a fertile development field.
A great precision is achieved due to the use of precise mathematical methods and algorithms that generate the artificial intelligence. Thus, absorbing and treating the abundance of travel data, the Neural Analyzer gets both, an effective learnig and a refined classification.
The appropriate offer of travel packages for travelers, briefly summarized above, is just one of the many uses of our system in travel agency activities, which could also use it - for example - to identify rentable offerings profiles, and occasional customers.
In the process of vehicle maintenance, the Neural Analyzer can bring an enormous ease in the diagnosis of motor and electrical defects, for example.
Among the main advantages for the use of this technology are: cost reduction by minimizing the necessary quantity of highly specialized labor; reduction of defect identification time; increased diagnosis accuracy; improvement of customer satisfaction by lowering the time for diagnosis and for performing the service.
By means of classification of the characteristics of the defects, with its detailed specification, together with the adequate and timely information on some characteristics of the vehicle and the driver, it is possible to infer with a high degree of certainty what’s the probable defect of the vehicle, as well as what should be done to achieve the best solution to the problem.
It began as a derivation from the software Analisador do Solo Urbano (Urban Soil Analyzer) the final product of the master dissertation in Applied Computing, defended by the civil engineer Carlos Cristiano Cabral in the Universidade Estadual do Ceará (UECE), a public university located in the northeast of Brazil, on may 31, 2010, when he had 44 years. The full text of the dissertation is accessible in www.dominiopublico.gov.br.
The original version of the Analisador do Solo Urbano software was created for didactic purposes, however with a conception widely connected with fiscal audit, in order to identify tax distortions related to the collection of the real estate tax (called IPTU) by the county of Fortaleza, area of wide domain and knowledge of the author due to the effective position he holds in the Municipal Finance Department (see resume below).
Carlos Cristiano Cabral is a civil engineer, master in Applied Computing, and auditor of taxes of the county of Fortaleza since 2004, where he administered the real estate department from 2006 to 2013. After that, he worked as an expert in the tax litigation department for two years, and headed from 2015 to 2019 the Active Debt Cell in the Attorney General's Office of the county of Fortaleza. He was also a professor at the University of Fortaleza (UNIFOR), where he coordinated the Computer Science course and taught subjects such as Computer Science, Data Processing, Numerical and Graphical Calculation and Programming Logic.
He is a lecturer and instructor in courses in Law, Computing and Public Administration, having published in this last area, the book "Vivências de um gestor público" (RDS, Brazil, 2012). He obtained the first place in the Sefin Prize of Municipal Public Finance, in the master’s dissertation category, edition 2010. He is the unique author of the software Analisador do Solo Urbano, with registration in the National Institute of Industrial Property (INPI) in 2016. By derivation, the Neural Analyzer was born from that software .
