Financial investment predictions
In various examples, interactive systems that use neural networks to determine financial investment predictions or recommendations are presented. Systems and methods are disclosed that determine financial predictions or recommendations associated with one or more investments using a neural network(s). The financial predictions may include a predicted movement of an investment (e.g., extremely down, down, preserved, up, extremely up, etc.), a predicted price of an investment (e.g., a future stock price, etc.), a specific investment for a user to buy/sell/trade, and/or so forth. In some examples, the systems and methods may include an interactive system(s), such as a dialogue system(s), that interacts with users to provide the financial predictions.
1 . A method comprising:
receiving, from a user device, input data representative of a request for information associated with an investment;
obtaining, based at least on the input data, financial data representative of previous values associated with the investment over a previous time period and event data representative of one or more financial events associated with the investment over the previous time period;
generating, based at least on a neural network processing the financial data using one or more first input layers that are trained to extract values, one or more first vectors associated with the previous values of the investment that are represented using a time interval and over the previous time period;
generating, based at least on the neural network processing the event data using one or more second input layers that are trained to extract one or more event words, one or more second vectors associated with the one or more financial events represented by the event data, the one or more second input layers being different than the one or more first input layers;
generating, based at least on the neural network processing the one or more first vectors and the one or more second vectors using one or more recurrent neural network layers, a first dense vector associated with the one or more first vectors and a second dense vector associated with the one or more second vectors;
generating, based at least on the neural network processing the first dense vector and the second dense vector using one or more subsequent layers that are trained to predict values, output data representative of future values for the investment that are represented using the time interval and over a future time period; and
causing an output of at least the future values using the user device.
2 . The method of claim 1 , further comprising:
obtaining prediction data representative of one or more predicted values associated with the investment; and
generating, based at least on the neural network processing the prediction data using one or more third input layers, one or more third vectors associated with the one or more predicted values,
wherein the generating the output data representative of the future values associated with the investment is further based at least on the one or more third vectors.
3 . The method of claim 1 , further comprising:
obtaining second event data representative of one or more second financial events associated with a second investment, the second investment being related to the investment; and
generating, based at least on the neural network processing the second event data using the one or more second input layers, one or more third vectors associated with the one or more second financial events,
wherein the generating the output data representative of the future values associated with the investment is further based at least on the one or more third vectors.
4 . The method of claim 1 , wherein:
the financial data is representative of the previous values at individual time intervals of the time interval and within the previous time period.
5 . The method of claim 4 , wherein:
the future values respectively correspond to the individual time intervals within the future time period.
6 . The method of claim 1 , further comprising:
obtaining second financial data representative of one or more second previous values associated with the investment over a second previous time period;
obtaining second event data representative of one or more second financial events associated with the investment over the second previous time period; and
determining, using the neural network and based at least on the second financial data and the second event data, a second future value associated with the investment.
7 . The method of claim 1 , wherein the neural network is trained, at least, by:
inputting training data into the neural network, the training data representative of one or more second previous values associated with the investment and one or more second financial events associated with the investment;
determining, using the neural network and based at least on the training data, a second future value associated with the investment;
comparing the second future value associated with the investment to an actual value associated with the investment, the actual value represented by ground truth data; and
updating, based at least on the comparing the second future value to the actual value, one or more parameters associated with the neural network.
8 . The method of claim 1 , further comprising:
determining the investment based at least on one or more of audio data representing user speech associated with the investment or a user profile that is associated with the investment,
wherein the user device is associated with at least one of the audio data or the user profile.
9 . The method of claim 1 , wherein the generating the output data represented of the future values for the investment comprises:
determining, based at least on the neural network processing the first dense vector and the second dense vector using the one or more subsequent layers, one or more regression scores associated with the investment; and
generating, using the one or more subsequent layers and based at least on the one or more regression scores, the output data representative of the future values of the investment that are represented using the time interval and over the future time period.
10 . The method of claim 1 , wherein:
the financial data represents one or more first input vectors corresponding to the previous values; and
the event data represents one or more second input vectors corresponding to the one or more financial events.
11 . One or more hardware processors comprising processing circuitry to:
receive, from a user device, input data representative of a request for information associated with an investment;
obtain, based at least on the input data, financial data representative of previous values associated with the investment over a previous time period and event data representative of one or more financial events associated with the investment over the previous time period;
generate, based at least on a neural network processing the financial data using one or more first input layers that are trained to extract values, one or more first vectors associated with the previous values of the investment that are represented using a time interval and over the previous time period;
generate, based at least on the neural network processing the event data using one or more second input layers that are trained to extract one or more event words, one or more second vectors associated with the one or more financial events represented by the event data, the one or more second input layers being different than the one or more first input layers;
generate, based at least on the neural network processing the one or more first vectors and the one or more second vectors using one or more recurrent neural network layers, a first dense vector associated with the one or more first vectors and a second dense vector associated with the one or more second vectors;
generate, based at least on the neural network processing the first dense vector and the second dense vector using one or more subsequent layers that are trained to predict values, output data representative of future values for the investment that are represented using the time interval and over a future time period; and
cause an output of at least the future values using the user device.
12 . The one or more hardware processors of claim 11 , wherein the processing circuitry is further to:
obtain prediction data representative of one or more predicted values associated with the investment; and
generate, based at least on the neural network processing the prediction data using one or more third input layers, one or more third vectors associated with the one or more predicted values,
wherein the output data representative of the future values associated with the investment is further generated based at least on the one or more third vectors.
13 . The one or more hardware processors of claim 11 , wherein the one or more hardware processors are comprised in at least one of:
a system for performing simulation operations;
a system for performing digital twin operations;
a system for performing deep learning operations;
a system implemented using an edge device;
an infotainment system of an autonomous or semi-autonomous machine;
a system implemented using a robot;
a system for performing conversational AI operations;
a system for generating synthetic data;
a system incorporating one or more virtual machines (VMs);
a system implemented at least partially in a data center; or
a system implemented at least partially using cloud computing resources.
14 . A system comprising:
one or more processors to:
receive, from a user device, input data representative of a request for information associated with an investment;
obtain, based at least on the input data, financial data representative of previous values associated with the investment over a previous time period and event data representative of one or more financial events associated with the investment over the previous time period;
generate, based at least on a neural network processing the financial data using one or more first input layers that are trained to extract values, one or more first vectors associated with the previous values of the investment that are represented using a time interval and over the previous time period;
generate, based at least on the neural network processing the event data using one or more second input layers that are trained to extract one or more event words, one or more second vectors associated with the one or more financial events represented by the event data, the one or more second input layers being different than the one or more first input layers;
generate, based at least on the neural network processing the one or more first vectors and the one or more second vectors using one or more recurrent neural network layers, a first dense vector associated with the one or more first vectors and a second dense vector associated with the one or more second vectors;
generate, based at least on the neural network processing the first dense vector and the second dense vector using one or more subsequent layers that are trained to predict values, output data representative of future values for the investment that are represented using the time interval and over a future time period; and
cause an output of at least the future values using the user device.
15 . The system of claim 14 , wherein the one or more processors are further to:
obtain prediction data representative of one or more predicted values associated with the investment; and
generate, based at least on the neural network processing the prediction data using one or more third input layers, one or more third vectors associated with the one or more predicted values,
wherein the output data representative of the future values associated with the investment is further generated based at least on the one or more third vectors.
16 . The system of claim 14 , wherein the one or more processors are further to:
obtain second event data representative of one or more second financial events associated with a second investment, the second investment being related to the investment; and
generate, based at least on the neural network processing the second event data using the one or more second input layers, one or more third vectors associated with the one or more second financial events,
wherein the output data representative of the future values associated with the investment is further generated based at least on the one or more third vectors.
17 . The system of claim 14 , wherein the financial data is representative of the previous values at individual time intervals of the time interval and within the previous time period.
18 . The system of claim 14 , wherein the one or more processors are further to:
obtain second financial data representative of one or more second previous values associated with the investment over a second previous time period;
obtain second event data representative of one or more second financial events associated with the investment over the second previous time period; and
determining, using the neural network and based at least on the second financial data and the second event data, a second future value associated with the investment.
19 . The system of claim 14 , wherein the neural network is trained, at least, by:
inputting training data into the neural network, the training data representative of one or more second previous values associated with the investment and one or more second financial events associated with the investment;
determining, using the neural network and based at least on the training data, a second future value associated with the investment;
comparing the second future value associated with the investment to an actual value associated with the investment, the actual value represented by ground truth data; and
updating, based at least on the comparing the second future value to the actual value, one or more parameters associated with the neural network.
20 . The system of claim 14 , wherein the system is comprised in at least one of:
a system for performing simulation operations;
a system for performing digital twin operations;
a system for performing deep learning operations;
a system implemented using an edge device;
an infotainment system of an autonomous or semi-autonomous machine;
a system implemented using a robot;
a system for performing conversational AI operations;
a system for generating synthetic data;
a system incorporating one or more virtual machines (VMs);
a system implemented at least partially in a data center; or
a system implemented at least partially using cloud computing resources.