View source for 17964084. ONE-HOT ENCODER USING LAZY EVALUATION OF RELATIONAL STATEMENTS simplified abstract (Oracle International Corporation)

=ONE-HOT ENCODER USING LAZY EVALUATION OF RELATIONAL STATEMENTS=


==Organization Name==

[[:Category:Oracle International Corporation|Oracle International Corporation]]


[[Category:Oracle International Corporation]]

==Inventor(s)==

[[:Category:FELIX Schmidt of Baden-Dattwil (CH)|FELIX Schmidt of Baden-Dattwil (CH)]][[Category:FELIX Schmidt of Baden-Dattwil (CH)]]

[[:Category:MATTEO Casserini of Zurich (CH)|MATTEO Casserini of Zurich (CH)]][[Category:MATTEO Casserini of Zurich (CH)]]

[[:Category:MILOS Vasic of Zurich (CH)|MILOS Vasic of Zurich (CH)]][[Category:MILOS Vasic of Zurich (CH)]]

[[:Category:MARIJA Nikolic of Zurich (CH)|MARIJA Nikolic of Zurich (CH)]][[Category:MARIJA Nikolic of Zurich (CH)]]

==ONE-HOT ENCODER USING LAZY EVALUATION OF RELATIONAL STATEMENTS - A simplified explanation of the abstract==


This abstract first appeared for US patent application 17964084 titled 'ONE-HOT ENCODER USING LAZY EVALUATION OF RELATIONAL STATEMENTS

==Simplified Explanation==

The abstract describes a method and storage media for training and implementing a one-hot encoder, which involves extracting unique categories, associating them with unique indices, and generating one-hot encodings. The process includes executing relational statements using a query optimization engine and encoding categorical features based on the encoder state.

* Explanation of the patent:
* During training, unique categories are extracted and associated with unique indices.
* One-hot encodings are generated for each unique category.
* Relational statements are executed using a query optimization engine.
* Execution of statements is postponed until needed, with optimizations implemented.
* Categorical features in a second training data set are encoded based on the encoder state.

== Potential Applications ==
The technology can be applied in various fields such as machine learning, data analysis, and natural language processing.

== Problems Solved ==
1. Efficient encoding of categorical features.
2. Optimization of query execution for better performance.

== Benefits ==
1. Improved data processing efficiency.
2. Enhanced performance in training and implementing one-hot encoders.

== Potential Commercial Applications ==
Optimizing data processing in industries such as e-commerce, finance, and healthcare.

== Possible Prior Art ==
Prior art may include existing methods for encoding categorical features and optimizing query execution in data processing systems.

== Unanswered Questions ==
=== How does the technology handle large datasets efficiently? ===
The efficiency of the technology with large datasets may depend on the scalability and optimization techniques used in the implementation.

=== What are the potential limitations of the query optimization engine? ===
The limitations of the query optimization engine, such as handling complex relational statements or specific data types, could impact the overall performance of the technology.


==Original Abstract Submitted==

A method and one or more non-transitory storage media are provided to train and implement a one-hot encoder. During a training phase, computation of an encoder state is performed by executing a set of relational statements to extract unique categories in a first training data set, associate each unique category with a unique index, and generate a one-hot encoding for each unique category. The set of relational statements are executed by a query optimization engine. Execution of the set of relational statements is postponed until a result of each relational statement is needed, and the query optimization engine implements one or more optimizations when executing the set of relational statements. During an encoding phase, a set of categorical features in a second training data set are encoded based on the encoder state to form a set of encoded categorical features.



[[Category:G06F16/2453]]
[[Category:G06F16/2458]]