Years ago, the problem in obtaining present or future (predictions) knowledge with data science was the lack of systems for storing and processing large amounts of information. This is not the case today. The efforts are therefore focused on how to analyze the data in order to extract true value. Deep Feature Synthesis helps in this task because it is an algorithm that automates automatic learning.
Machine learning is the set of processes whereby an algorithm is capable of making predictions from data, and where the result of that projection enhances the machine’s own learning to improve its predictions. Machines learning from their mistakes and successes. It is therefore a specific branch of artificial intelligence that is applied in fields as diverse as banking to detect fraud processes, health for hospital management, or retailing for optimized price calculation.
The two creators of the algorithm are two prominent members of the MIT‘s Computing Science and Artificial Intelligence Laboratory in Cambridge. Their names: James Max Kanter and Kalyan Veeramachaneni. They both presented the project in a document entitled ‘Deep Feature Synthesis: Towards Automating Data Science Endeavors’ (PDF), in which they summarize the characteristics of their ‘creature’.
Deep Feature Synthesis does exactly what its own name announces. It is an algorithm capable of automatically creating features between sets of relational data to synthesize automatic learning processes. The algorithm applies mathematical functions to the data sets in the source field and transforms them into new groups with new and deeper features.
In this process of evolution of the source data, one can begin with some information variables referring to the gender or age and at the end of the process applied by Deep Feature Synthesis have features that will make it possible to make other deeper calculations as percentages.
Deep Feature Synthesis and the Gaussian copula
This automatic machine learning process is perfected, according to its creators, thanks to the probability theory of the Gaussian copula. Many of the stages in an automatic learning process have parameters that, looking for an appropriate result, require a tuning process. The less tuning is used, the less predictive the model will be.
Small variations in that improvement can result in chaos. The large amount of combinations of parameters turns any minimum deviation at the beginning into a huge error at the end. In economic predictive models, this means billions of dollars. The Gaussian copula helps the algorithm’s creators to model the relationship between the options of the parameters and the performance of the entire model. From there, a decision is made as to the best parameters for optimizing the result.
The Gaussian copula was the statistical model used to avoid a major credit crisis like the one in 2008 and it obviously did not succeed. This method was used in VAR (Value at Risk) analyses, that measure the losses that the market could sustain in normal conditions with a confidence level of 95%. In other words, that an investor with a one million euros portfolio can only lose 25,000 euros each of the 20 days (1/20=5% remaining from the confidence level). The day when the great international crisis broke out in 2008, those values soared outside the margins.
Implementation of Deep Feature Synthesis
The Deep Feature Synthesis algorithm and its Data Science machine are implemented on top of an MySQL database, using InmoDB as the engine for the tables, an open source data storage solution for this type of relational database. InmoDB replaces the previous table technology for MySQL and MyISAM. It is more reliable, more consistent, more scalable and, therefore, it offers more performance.
All the data sets with which Deep Feature Synthesis works are converted into the data scheme used by MySQL. The calculation logic, the management and the handling of the features of all this information is done through the programming languagePython, the most widely used syntax for the design and configuration of Data Science processes.
Why did the creators of Deep Feature Synthesis use a relational database like MySQL? Because the algorithm’s requirements and the way in which the data in this type of database are sorted are similar. The Data Science machine used by the algorithm’s creators implements AVG (), MAX (), MIN (), SUM (), STD () and COUNT () type functions. It also adds others for another type of operations with the data, such as length () or WEEKDAY () and MONTH () to convert the dates into the days of the week and the month when they occurred.
The use of functions, plus the creation of filters, enable the algorithm to address two really important matters in predictive models:
● Application of functions only to the cases in which a given condition is taken to be true. Without data filtering it is impossible.
● Construction of time interval values, with limits above and also below, based on a time date.
This makes it much easier to optimize database queries.
The three processes of Deep Feature Synthesis
The Data Science based on Deep Feature Synthesis uses the three usual processes with the data to prepare predictive models:
● Data pre-processing: the preliminary work with the data is essential before considering automatic learning work. The parameters need to be reviewed in order to reject, for example, null values.
● Selection of features and reduction of dimensionality: the algorithm generates a large amount of features for each entity and a preliminary selection and reduction task is necessary.
● Modeling: decision trees are used for data modeling.
An API is a very useful mechanism that connects two pieces of software equipment to exchange messages or data in a standard format such as XML or JSON. Thus, it becomes an instrument that can be used to search for revenue, open the doors to talent or innovate and automate processes.
APIs can be a great support when automating business processes Companies, often with a focus on SMEs, spend too many man-hours on time-consuming business processes, thereby making mistakes that a machine would never make. How can business process automation (BPA) help these companies? Is it possible to make use of APIs for BPA? What is […]
Ecommerce has continued to grow steadily in Spain, except during the pandemic, which has already been overcome in terms of online shopping. Ecommerce has been making inroads among the Spanish for over two decades. In 2000, it was a marginal and niche activity. Now it is almost universal. Almost all Spaniards with internet access shop online […]
Please, if you can't find it, check your spam folder
×
The email message with your ebook is on the way
We have sent you two messages. One with the requested ebook and one to confirm your email address and start receiving the newsletter and/or other commercial communications from BBVA API_Market
×
PROCESSING OF PERSONAL DATA
Who is the Data Controller of your personal data?
Banco Bilbao Vizcaya Argentaria, S.A. (“BBVA“) with registered address at Plaza de San Nicolás 4, 48005, Bilbao, España and Tax ID number A-48265169 . Email address: contact.bbvaapimarket@bbva.com
What for and why does BBVA use your personal data for?
For those activities among the following for which you give your consent by checking the corresponding box:
to receive newsletter from BBVA API_Market through electronic means;
to send you commercial communications, events and surveys relating to BBVA API_Market to the e-mail address you have provided.
For how long we will keep your data?
We will keep your data until you unsubscribe from receiving our newsletter or, if applicable, the commercial communications, events and surveys to which you have subscribed. Whether you unsubscribe or whether BBVA decides to end the service, your details will be deleted.
How can I unsubscribe to stop receiving newsletters and/or communications from BBVA API_Market?
You can unsubscribe at any time and without need to indicate any justification, by sending an email to the following address: contact.bbvaapimarket@bbva.com
To whom will we communicate your data?
We will not transfer your personal data to third parties, unless it is mandatory by a law or if you have previously agreed to do so.
What are your rights when you provide us with your information?
You will be able to consult your personal data included in BBVA files (access right)
You can modify your personal data when they are inaccurate (correction right)
You may request that your personal data not be processed (opposition right)
You may request your personal data be deleted (suppression right)
You can request a limitation on the processing of your data in the allowed cases (right of limitation of processing)
You will be able to receive, in electronic format, the personal data you have provided to us, as well as to transmit them to another entity (portability right)
You are responsible for the accuracy of the personal data you provide to BBVA and to keep them duly updated. If you believe that we have not processed your personal data in accordance with regulations, you can contact the Data Protection Officer of BBVA at the following address dpogrupobbva@bbva.com.
You can find more information in the “Personal Data Protection Policy” document on this website.
×
PROCESSING OF PERSONAL DATA
Who is the Data Controller of your personal data? Banco Bilbao Vizcaya Argentaria, S.A (“BBVA“), with registered address at Plaza de San Nicolás 4, 48005, Bilbao, España, and Tax ID No. A-48265169. Email address:contact.bbvaapimarket@bbva.com
What for and why does BBVA use your personal data for?
For the execution and management of your request, specifically, download the requested e-book/s.
BBVA informs you that, unless you indicate your opposition by sending an email to the following address: contact.bbvaapimarket@bbva.com, BBVA may send you commercial communications, surveys and events related to products and/or services of BBVA API Market through electronic means.
For how long we will keep your data?
We will keep your data as long as necessary for the management of your request, and to receive commercial communications, events and surveys. BBVA will keep your data until you unsubscribe to stop receiving our newsletters or, where appropriate, until the end of the service. Afterwards, we will destroy your data.
How can I unsubscribe to stop receiving newsletters and/or communications from BBVA API Market?
You can unsubscribe at any time and without need to indicate any justification, by sending an email to the following address: contact.bbvaapimarket@bbva.com
To whom will we communicate your data?
We will not transfer your personal data to third parties, unless it is mandatory by a law or if you have previously agreed to do so.
What are your rights when you provide us with your information?
You will be able to consult your personal data included in BBVA files (access right)
You can modify your personal data when they are inaccurate (correction right)
You may request that your personal data not be processed (opposition right)
You may request your personal data be deleted (suppression right)
You can request a limitation on the processing of your data in the allowed cases (right of limitation of processing)
You will be able to receive, in electronic format, the personal data you have provided to us, as well as to transmit them to another entity (portability right)
You can exercise before BBVA the aforementioned rights through the following address: contact.bbvaapimarket@bbva.com
You are responsible for the accuracy of the personal data you provide to BBVA and to keep them duly updated.
If you believe that we have not processed your personal data in accordance with the regulations, you can contact the Data Protection Officer at the following address: dpogrupobbva@bbva.com
You can find more information in the “Personal Data Protection Policy” document on this website.
Banco Bilbao Vizcaya Argentaria, S.A. owner of this portal uses cookies and/or similar technologies of its own and third parties for the purposes of personalization, analytics, behavioral advertising or advertising related to your preferences based on a profile prepared from your browsing habits (e.g. pages visited). If you wish to obtain more detailed information, consult our Cookies Policy.
Cookie settings panel
These are the advanced settings for first-party and third-party cookies. Here you can change the parameters that will affect your browsing experience on this website.
Technical Cookies (required)
These cookies are used to give you secure access to areas with personal information and to identify you when you log in.
Name
Owner
Duration
Description
gobp.lang
BBVA
1 month
Language preference
aceptarCookies
BBVA
1 year
Configuration Accepted Cookies
_abck
BBVA
1 year
Helps protect against malicious website attacks
bm_sz
BBVA
4 hours
Helps protect against malicious website attacks
ADRUM_BTs
Salesforce Marketing Cloud
Session
Required for monitoring of the service, inherent to SFMC
ADRUM_BT1
Salesforce Marketing Cloud
Session
Required for monitoring of the service, inherent to SFMC
ADRUM_BTa
Salesforce Marketing Cloud
Session
Required for monitoring of the service, inherent to SFMC
ADRUM_BT
Salesforce Marketing Cloud
Session
Required for monitoring of the service, inherent to SFMC
xt_0d95e
Salesforce Marketing Cloud
Session
Remember user preferences (if any)
__s9744cdb192d044faa1bf201d29fafd1e
Salesforce Marketing Cloud
Session
Remember user preferences (if any)
wpml_browser_redirect_test
WPML
Session
Text translation in the portal
wp-wpml_current_language
WPML
24 hours
Text translation in the portal
They are used to track the activity or number of visits anonymously. Thanks to them we can constantly improve your browsing experience
Your browsing experience is constantly improving.
With your selection, we cannot offer you a continuously improved browsing experience.
Name
Owner
Duration
Description
AMCV_***
Adobe Analytics
Session
Unique Visitor IDs used in Cloud Marketing solutions
AMCVS_***
Adobe Analytics
2 years
Unique Visitor IDs used in Cloud Marketing solutions
demdex (safari)
Adobe Analytics
180 days
Create and store unique and persistent identifiers
sessionID
Adobe Analytics
Session
Launch's internal cookie used to identify the user
gpv_URL
Adobe Analytics
Session
Adobe Analytics plugin: getPreviousValue Capture the value of a certain variable in the following page view, in this case the prop1
gpv_level1
Adobe Analytics
Session
Cookie used to store the DataLayer levl1 of the previous page.
gpv_pageIntent
Adobe Analytics
Session
Cookie used to store the pageIntent of the previous page.
gpv_pageName
Adobe Analytics
Session
Cookie used to store the pagename of the previous page.
aocs
Adobe Analytics
Session
Cookie that stores the first values collected at the beginning of a process.
TTC
Adobe Analytics
Session
Cookie used to store the time between the App Page Visit event and the App Completed event.
TTCL
Adobe Analytics
Session
Cookie used to store the time between the LogIn event and App Completed.
s_cc
Adobe Analytics
Session
Determine if cookies are active
s_hc
Adobe Analytics
Session
Cookie used by Adobe for analytical purposes
s_ht
Adobe Analytics
Session
Cookie used by Adobe for analytical purposes
s_nr
Adobe Analytics
2 years
Determine the number of user visits
s_ppv
Adobe Analytics
Permanent
Adobe Analytics plugin: getPercentPageViewed Determine what percentage of the page a user views
s_sq
Adobe Analytics
Session
ClickMap/ActivityMap features
s_tp
Adobe Analytics
Session
Cookie used by Adobe for analytical purposes
s_visit
Adobe Analytics
2 years
Cookie used by Adobe to know when a session has been started.
They allow the advertising shown to you to be customized and relevant to you. Thanks to these cookies, you will not see ads that you are not interested in.
The advertising is customized to you and your preferences.
Your choice means you will not see customized ads, only generic ones.
Name
Owner
Duration
Description
OT2
VersaTag
90 days
VersaTag Cookie used to store a user id and the number of user visits.
u2
VersaTag
90 days
VersaTag Cookie where the user ID is stored
TargetingInfo 2
MediaMind
1 year
Cookie that serves to assign a unique random number that generates MediaMind.
These cookies are related to general features such as the browser you use.
Your experience and content have been customized.
With your selection, we cannot offer you a continuously improved browsing experience.
Name
Owner
Duration
Description
mbox
Adobe Target
9 days
Cookie used by Adobe Target to test user experience customization.
×
Looks like you’re browsing from Mexico, so let’s show you the custom content for your
location. Change
Looks like you’re browsing from Spain, so let’s show you the custom content for your
location. Change
Select a country
In order to access the private area and corresponding sandbox, select the country of the APIs you want to use.