Adapt Machine Learning Algorithm for overridden decisions
We have insurance data of over 10 years. There are underwriting rules for the data, which result in two possible outcomes: Approve or Reject.
We want to have a Machine Learning Algorithm to learn these rules and predict the outcome for the future cases, which is all fine. BUT, if the socio-economic condition changes, the underwriter will override the ML decision manually. It is expected for the system to adapt accordingly and behave in that fashion for the upcoming applications.
Is there any possible way(s)?
See also questions close to this topic
-
how to detect mathematical calculations from a photo of math workbook and determine if it is correct
Actually i have trained some models from faster_rcnn_resnet101,the dataset contains about 10k images(each is a sheet from some math workbook),but the detection ranges are not very accurate,many numbers are cut off horizontally or vertically which makes them recognized as other numbers
For example recognize 48÷6=8 as 48÷6=0,because 8 is handwritten and it's a little bit higher above 48 and 6. Or 12÷2=6 as 2÷2=6 in which the first 1 is totally cut off
So i'm wondering if the faster_rcnn_resnet101 model is not appropriate for this kind of situation or maybe it's the problem of my training dataset(i can make sure every image in the dataset is well tagged)
If faster_rcnn_resnet101 shouldn't be used here,what kind of other object detection models can i use.
-
Machine Learning Platform for Sales Team
I built a model that predicts how potential sales opportunities would perform. The goal is for our sales team to be able to plug in different numeric and categorical variables that would then give them the predicted sales value.
I’m currently using R Shiny but was wondering what other platforms can do this or what others would recommend.
-
Matrix shapes difference between lectures
I'm currently learning deep-learning from two lectures. What gets me confused is that there is a notation difference between two lectures when they shape an input matrix X.
In Coursera's lecture, they make a matrix X in shape of (number of features, number of samples), so that they stack the samples vertically. Otherwise, the other lecture stacks the samples horizontally, so that every row represents one sample.
What makes this difference and which one should I follow?
-
How to make user input with voice gtts
I am creating a simple pizza app with voice inputs. Before creating this post i have searched a lot and didn't find anything useful.
The fiddle doesn't run the code appropriately so don't run it over there.
The issue:
When i try to use "command" in my "input" (e.g.
count = 0 def talkToMe(audio): global count print(audio) text_to_speech = gTTS(text=audio, lang='en-us') text_to_speech.save(f'speech{count%2}.mp3') mixer.init() mixer.music.load(f'speech{count%2}.mp3') mixer.music.play() count += 1 def myCommand(): r = sr.Recognizer() with sr.Microphone() as source: print('Ready...') r.pause_threshold = 1 r.adjust_for_ambient_noise(source, duration=1) audio = r.listen(source) try: command = r.recognize_google(audio).lower() print('You said: ' + command + '\n') #loop back to continue to listen for commands if unrecognizable speech is received except sr.UnknownValueError: print('Your last command couldn\'t be heard') command = myCommand(); return command def assistant(command): def pick_or_deli(): global delivery global customer_name global customer_telephone delivery = input("pickup - pick up / delivery - delivery:" + command) delivery = delivery.upper() if delivery == "DELIVERY": while running == True: ..............................
i am simply getting:
Don't pay attentio on lowercase "delivery". Even if i change this line:
if delivery == "DELIVERY":
to this:
if delivery == "delivery":
It doesn't work. Even if i hit enter. The system doesn't recognize it as an input.
Any solutions/suggestions on this would be great.
-
Statistical Modeling/Machine Learning for the particular goal?
The data snippet is related to plant growth automation system where the readings are recorded from different sensors like (light, temperature, soil moisture) etc. Dataset Columns are
The idea is to monitor the plant vegetation health. My doubt is:-
1) To start with what data analysis/unsupervised tasks can be done that can help towards identifying factors for the goal to measure the plant health.
2) how & what can we assign as outcomes 'Y' for this data, if supervised learning is to be done?
3) What assessment criteria can be used related to vegetation indexes?
4) What vegetation/cultivation measures can be used?
-
Problem with the word 'machine' in the definition of machine learning by Mitchell
The definition : A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E.
Here, we talk about what it means for a program to learn rather than a machine, and a program and a machine aren't equivalent so how do we justify calling it 'machine' learning?
I have gone through similar questions ex: What is machine learning? but it doesn't address this aspect.