Digital Transformation in the fight against Coronavirus
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
Title |
| |
Subtitle |
| |
Title of Series | ||
Number of Parts | 130 | |
Author | ||
License | CC Attribution - NonCommercial - ShareAlike 3.0 Unported: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal and non-commercial purpose as long as the work is attributed to the author in the manner specified by the author or licensor and the work or content is shared also in adapted form only under the conditions of this | |
Identifiers | 10.5446/49912 (DOI) | |
Publisher | ||
Release Date | ||
Language |
Content Metadata
Subject Area | ||
Genre | ||
Abstract |
|
00:00
Transformation (genetics)Digital signalMathematical analysisGoodness of fitState of matterDigital signal processingAerodynamicsInformationFrequencyComputer animation
00:35
Mathematical analysisMachine learningState of matterService (economics)Virtual machineProfil (magazine)Physical lawDistanceObject (grammar)Latent heatInsertion lossComputer animation
02:09
Mathematical analysisSystem identificationCodeEndliche ModelltheorieCommunications protocolMathematical analysisVirtual machineInteractive televisionComputer animation
03:01
Computer virusArtificial intelligenceMathematical analysisCall centreMathematical analysisWaveState of matterVirtual machineCurveMeasurementDependent and independent variablesEndliche ModelltheorieArtificial neural networkInteractive televisionMachine learningLatent heatDistanceComputer animation
04:06
Digital object identifierFrequencyInverse elementTerm (mathematics)Group actionLocal GroupDivision (mathematics)SpacetimeDiagramVoronoi diagramPartition (number theory)Point cloudLogicData structureService (economics)State of matterOpen setService (economics)Term (mathematics)LogicEndliche ModelltheorieFrame problemMoment (mathematics)Server (computing)Connectivity (graph theory)Open setEmpirical distribution functionInteractive televisionGene clusterFrequencyWeb 2.0Mobile appFood energyDecision theoryPower (physics)Point cloudLogic synthesisVisualization (computer graphics)Mathematical analysisLatent heatWordData storage deviceWeb applicationComputer animation
07:34
Universe (mathematics)Client (computing)Communications protocolEndliche ModelltheorieLatent heatWordConstructor (object-oriented programming)Data storage deviceService (economics)Point cloudOperator (mathematics)Interactive televisionGene cluster
08:44
Service (economics)Computer hardwareMereologyNeighbourhood (graph theory)Service (economics)WordCommunications protocolCountingSystem administratorLatent heatPoint cloudServer (computing)Hydraulic motorComputer animation
09:26
Vector spaceGene clusterEndliche ModelltheorieWordProduct (business)Cartesian coordinate systemOperating systemSlide ruleMessage passingGroup actionVariable (mathematics)Point cloudType theoryNetwork topologyDecision theoryInteractive televisionDistribution (mathematics)Virtual machineForestSocial classLibrary (computing)Software testingPhysical systemData storage devicePresentation of a groupSound effectNumberSet (mathematics)Computer programmingService (economics)Arithmetic meanGoogolLatent heatAlgorithmData conversionRoundness (object)Empirical distribution functionFitness function4 (number)Communications protocolDifferent (Kate Ryan album)XMLComputer animation
17:47
PredictabilityDigital signal processingCurveCodePresentation of a groupContent (media)Computer animation
18:13
Meeting/Interview
Transcript: English(auto-generated)
00:06
Okay, good afternoon for everyone. It's a pleasure to stay here at Aero Pi Tom and have the opportunity to share our experience and government of the state of the Guayas with the pandemic and how digital transformation
00:25
help us to give information, to give for the society a better public policies in this difficult period, yeah? So today we discuss about the COVID-19 and the impact of the demand for this service
00:44
in our channels here in the state. So we have a huge problem. So COVID-19 increase the demand for complaints and our channels of the embarrassment here and the government of the state.
01:02
So the people stay very angry. So here in Brazil, we have many loss about social distance and people have doubts about what's need to be, what's need to stay open or need to stay close, what service can be continue work or not continue work.
01:24
So we have this idea and these doubts impact in our channels of the attendances by this. So don't have infrastructure, we don't have people to support these. So if one artificial intelligence, we can improve,
01:45
we can give support for the sentences to give a good attendance for each demand that people want. So our main object is how to apply machine learning to identify interactions profile in the business channel
02:03
and the state of the APIs and give support and the specific support for everyone. So for these, before that we apply codes about machine learning, codes about Python, codes about a specific model,
02:21
we need to identify and understand the problem. So first it's, we need to understand and analyze how the population, the cities interact with our channels. After we, in the next step, we build a descriptive analysis for these interactions. So in that other step, we build a text analysis
02:44
and use a clustering techniques to identify personas, to identify ideas, to give support. So we, with these personas, we had the opportunity to build protocols, to build a recommendation to give any specific attendance.
03:02
So this is our peak. So in the pandemic, we have peak, we have curves and here in state of the APIs, we have peaks and we have curves too. So it's the normal peak and the normal wave of the demand for our channels to complain.
03:21
After the measures of the social distancing, we increase for these demands. So we don't have the infrastructure, we don't have attendance in the call center to support this increase of the demand.
03:40
So we need to build them at a specific tools to response for these problems. So with our artificial intelligence, with machine learning, we can give a good response for these. To improve our channels and to improve the interactions,
04:02
we build a text analysis with artificial model. So it's our simple frame of the interaction. So we have the manifestations, we apply NLP, we apply a k-mean clusters to identify this persona,
04:21
we apply a text model classifications to identify these personas, we deploy of these models in a web app and we build a data visualization to take the better decision in the moment that we want.
04:45
So for these infrastructures of this model, we use TF-DF model. TF-DF model, it's the basic model when you discuss about text analysis. We consider the term, fragrance in the entire document
05:01
of the text that we consider. We use a k-mean clusters to identify the subjects of the interest and we have a big picture of these interactions and the big picture of these analysis.
05:20
Of course, this model is in Portuguese. This tech cloud is in Portuguese, but we have opportunity to identify some stronger words like employers, like companies, like the district, like the servers, like the other companies, like the government. So we have a big picture of this idea.
05:42
And we have a logic, the background of these interactions. We have common citizens that complain about one company. We have the employer that complain about a service
06:01
that working and in some moments, the servers didn't work in this moment and we have complaints about that the activity need to be working or not. So it's the main logic in these interactions. And after we have tags for the clusters.
06:25
So we have a cluster of the interactions that it's running activity and to employ that agglomeration in a specific service. We have citizens that complain about open servers. We have that employee that request about protection.
06:43
We have servers of the entertainment that open. We have closed door companies that is still working besides the decrees. We have decoration stores that's open. We have open bars that work. This is important because we don't have energy.
07:03
We don't have power. We don't have infrastructures to attendance everyone. So these personas give for us the idea that we need to focus our energy and a specific service.
07:22
Like, okay, open bar, it's not a service that we don't agree that open bar did work. So we act in a specific service here. So we have a clustering that we have operation activities
07:41
and employee and agglomerations. And we have specific ideas that was reported here. So we have work constructions, we have universities, we have IT companies, and we have any specific idea. For example, these specific service, this is specific store here in our district have made a more 20 requests in one hour.
08:03
So we have and build protocol, specific protocols for each one, this cluster interaction. So here we have any specific cloud for these ideas. So we have a strong word.
08:20
So employers, we have companies and we build a nice specific model to understand how this word was related. So the word employee and it's very close to risk to put these workers in risk, to put these clients in risk and we put these workers to work in risk.
08:45
We have citizens that complain about open service. So we have many doubts. Many of the service, it's allowed to open, but the citizens have doubt about this. So we resolve and solve these with specific protocol and specific ideas.
09:03
So we have administrative service, park, car wash, we have hotels and motels. We don't know if the service need to be working. So we have any specific word count. We need specific, I can see the word working.
09:21
It's very strong and working normally and how it's related about the idea. So I hope that you can see my Google collapsed. And the idea here, it's to try to show how we build this model and details.
09:43
We use any specific libraries. We use NLT key, use JSON and we use another libraries. This is our data set of the operation system. So we have protocol, we have manifestation, we have local effect and we analyze
10:01
any specific here manifestation that it's the text and the message that the citizens want to the government acting for doubts or other things. So we process this text. So we analyze, for example, a bigger model
10:22
and give a good insights for us. We have the opportunity to identify specific service. So we have commerce of to set a ice cream, for example, that's still working. We have the specific words that was related. We have this our word cloud about these topics.
10:46
And we start to process this text to understand how the word or the specific word interact. So for example, the word working, it's very related with the company is still working,
11:04
the company is still working, the company is still working normally. So we have the opportunity to understand these interactions. So here that I show for you in the slide,
11:24
we import the TF and EDF vectorize. We build and normalize and create a vector of this text. So we take this text here of the manifestation. So that's the text that the people typing
11:40
and transform these in a vector with this library. After this, we have the opportunity to apply technique clusters to identify interest and interaction. So we define eight clusters for each one,
12:01
we fit this model, and we have a clusters for the interest. So this is the distribution by cluster. So each number represent a cluster and the hard work is to tag and to give a meaning for each cluster.
12:21
And this is that we present for everyone in the slide and the clusters of the interaction. So after this, we have defined the cluster. So we define the group of the interest.
12:41
So the idea with this tool is to try to predict by the text what cluster each test was situation with the class. So this algorithm, this program here,
13:01
it's working with this. So here we apply techniques of the machine learning. So here we have, for example, the clusters, we have the model that we train. So we create a model of the classification. The idea here is to predict this cluster, this target variable with base of the manifestation.
13:24
So when the on the sentence type this, what's the cluster that we have the more probability that the text to stay, okay? Here we have the clusters that we define. We tokenize this text to define
13:44
this classification model. We define a target variable and we define a predict variable that was the vectorized text. So we try to experiment some models about classification
14:01
so we try to first with a multinomial model with the predict and with the target. So we don't have a good accuracy and 56%. We try to predict with a Gaussian model. We don't have a good accuracy too, we have 7%.
14:22
We try to predict with a decision tree model. We have an increase, we have 66% in the precision, but the best model that was performing the model of the classification, it's the random forest.
14:43
So we choose the random forest to predict with base in a text what's the group of the interaction that the text was situated, yeah? So we hear the scores of the text, yeah?
15:01
And after this, we export this model of the random forest to deploy this application in a production model. So here we convert this. We predicted this model with base of this text.
15:22
We have and create a distribution of the probability of this text was situation and if cluster, yeah? So here we have, for example, this message and the probability of this message stay and if each one cluster.
15:41
So this message, we have a good accuracy that was pretense for the cluster five and we export this model to has API to access for the system. In the end, we have a good idea. This is a model, a system that we deploy in the reluco.
16:04
So this is a type of the manifestation. So the idea of the message, it's important in Portuguese model for this is in Portuguese. So the district of the Agua Zlingas, the commercials is to work normally with a lot of agglomeration
16:21
and don't respect the social distance and remember that districts, it's so crowd. So the idea here, what's the group of the interest that this message was situated? With this cluster, with this model of the classification and with this model of the segment of the text,
16:43
we have a model, for example, again, okay. We have this message, it's too close to activities that still working with employers and agglomerations.
17:04
Another, we have, for example, a different message. For example, the store of the coffee is still working normally today and today it's Sunday and we don't have protections for our employees. What's the cluster that's the message, what situation?
17:20
So we have the clusters that the employees asking for protections. So it's important because give for us a specific target that how the government need to fighting or how the government need to activate
17:40
with the manifestations with the complaints. So this is the idea of the presentation to show, to share our experience and how digital transformation can help us in the pandemic of the COVID-19.
18:02
So we have our contents, I have in the Discord and we have a GitHub with the code and we can discuss another opportunity. Okay, thank you. Thank you, Bruno. Thank you very much, that was super nice.
18:20
So I'm going to play some songs. Thank you.