Artificial Intelligence (AI) and big data are a recen phenomen in a world in which activities continuously generated data, storing them in digital format. So, what is the secret of success? Build a data culture, this is a key to success of data valorisation and strategic steps, so we live the transition to the digitalized era.
Yuval Noah Harari (Sapiens: A Brief History of Humankind. 2105, p. 41) wonder “what happened in the Cognitive Revolution” and this author concludes that:
- “The ability to transmit larger quantities of information about the world surrounding Homo sapiens
- The ability to transmit larger quantities of information about Sapiens social relationships
- The ability to transmit information about things that do not really exist, such as tribal spirits, nations, limited liability companies, and human rights”
Harari (Sapiens: A Brief History of Humankind. 2015, p. 41) emphasizes that, “The Cognitive Revolution is accordingly the point when history declared its independence from biology.”
As the title says, "Another Cognitive Revolution" is because of Artificial Intelligence, so I think that Artificial Intelligence is really a cognitive revolution, not just a tool! that’s quite paradoxical, so this revolution is based on biology; the heart of neural networks as is shown below:
Fig. 1: Biological
Neuron.
Artificial Intelligence is quickly becoming more than just a tool. It’s begun to be the go-to option for Project Management everyday needs.
So, focus on Project Managment, we wonder how to Estimate Software Development Project In Man-Hours, which is the amount of work performed by the average worker in one hour. It’s essential to provide general estimate that’s based on our experience with similar projects and our historical data, then every Project Manager can make better estimations based on data.
Software development are calculated based on the parameters estimated, and depend on Software solutions design. Anyway people have the misapprehension that we can press a "magic button" that could do anything.
We need to re-evaluate and re-calculate all again, in order to adjust the project. So, that involves the implementation of a change control system, in which sophisticated and smart machines struggle to interpret deviation from predictions.
So, now let’s explain in high level what is happening in a neural network. In fact, we focus on forward propagation part of the network. In which we multiply each input by correspondent weight coefficient of each connection between the unites/nodes, and putting that through an activation function (see below).
Fig. 2: Procedure to Final Output
Fig. 3: The
network and parameters – illustrative example.
Artificial Intelligence is quickly becoming more than just a tool. It’s begun to be the go-to option for Project Management everyday needs.
So, focus on Project Managment, we wonder how to Estimate Software Development Project In Man-Hours, which is the amount of work performed by the average worker in one hour. It’s essential to provide general estimate that’s based on our experience with similar projects and our historical data, then every Project Manager can make better estimations based on data.
Software development are calculated based on the parameters estimated, and depend on Software solutions design. Anyway people have the misapprehension that we can press a "magic button" that could do anything.
We need to re-evaluate and re-calculate all again, in order to adjust the project. So, that involves the implementation of a change control system, in which sophisticated and smart machines struggle to interpret deviation from predictions.
So, now let’s explain in high level what is happening in a neural network. In fact, we focus on forward propagation part of the network. In which we multiply each input by correspondent weight coefficient of each connection between the unites/nodes, and putting that through an activation function (see below).
Once you have all this, now, you should apply steps as follows:
1. Multiply all m features in input X by m specific weights Z1 = X W(1)
2. Apply the activation function a1 = f(Z1)
3. We use hidden output as input data that has n features, multiply this new output by 1 set of n weights (w1, w2, …, wn): Z2 = X W(2)
4. Continue until reach the output layer.
5. Get your final output that represents our predictions: Y.
My thoughts are that data should be used to create a competitive advantage. The lack of comprehensive data could provide chaotic information, instead of valuable insights. So, Artificial Intelligence, with a mechanism to learn, advance the use of data, enriching decision support. The next step into the future.
Hope it helps!
Joan Martí.
