Week 06 Flashcards
What does GIS Stand for?
Geographic Information Systems
What are GIS’s? with examples
Deal with Geospatial Data - data that identifies the geographic location of features. Typically coordinates
Ex of spatial databases available:
Ordanace Survey
openstreetmap
Google Maps
What do spatial techniques do
Identify trends not obvious from other forms of analysis.
Ex: John Snows work in 1854 on the cholera spreading and identifying the water pump at the epiccentre of the disease.
What are the different types of spatial autocorrelation
Positive spatial autocorrelation
No spatial autocorrelation
Negative spatial autocorrelation
What is choropleth mapping
Display of attribute (non spatial) data associated with spatial entities.
Easier to see paterns.
What are spatial operations
Identify spatially related objects Ex: Beside, enclosed within, within a radius of
What is a Buffer?
Buffers are bands around spatial objects.
e.g. buffer around an airport might identify areas subject to noise
Why is buffer method not very sophisicated?
Circles are not overly accurate - as often drawn as the crow flies rather than travel distance.
Who are the traditional GIS users & list the types of spatial decision making they use?
Traditional users:
Utilities- EBS
Forestry- Coillte
Government
Spatial decision making:
location analysis
transport/logistics decisions
Marketing
Insurance
What are Insurance Uses of GIS
- Existing policies. Show who and where has policies, what is protected, and for which sum.
- Firmographics. This type shows which
businesses are located side by side. e.g. is a bookshop near a fire station or a firework shop. - Historical loss information. Spatial record of all losses for the entire observation period.
- Location risk index. For customers, each Location features a specific risk index.
- Natural disasters. fires, floods, tornadoes, earthquakes, etc
What is AI?
Behaviour which if performed by a human would be considered intelligent.
What are the three task domains of AI? - What will be the most difficult for computers to adjust to
Expert Tasks, Mundane tasks (most difficult for computers) and formal tasks.
Give an example of expert tasks as part of the task domains of AI?
Engineering, medical diagnosis, financial analysis
Give an example of mundane tasks as part of the task domains of AI?
Perception, robotics, common sense reaosning
Give an example of formal tasks as part of the task domains of AI?
Natural Language, Mathematics, Games
Explain AI techniques under accessing existing knowledge
- Expert systems
– performs a task that would otherwise be performed by a human expert
– reasoning approach
– “storage” of pre-established knowledge - Case Based Reasoning
– What did we do last time?
Explain AI techniques under creating new knowledge idea
Machine learning - system itself updates knowledge.
Explain what is an expert system?
A computer program that emulates the behaviour of human experts who are solving real problems associated with a particular domain of knowledge.
Not used by experts, but stores expert knowledge. – Only in very specific detailed areas
What are the benefits of expert systems?
- faster decisions – foreign exchange dealing
- better decision quality
- capture of scarce expertise – limited number of experts and an expert may not be in correct place
- integration of several opinions
- cheaper control devices
- spread of knowledge
- reliability
What are some problems with expert systems?
- Getting knowledge from expert is hard – Too busy or they may not understand the nature of the information required
- System is frozen in time – No systematic updating
- Expert rules can be inflexible
Explain what is case-based reasoning and what are steps required for CBR.
Solving problems automatically by doing what you did last time - but no problem is exactly the same.
CBR learns so long as main problem does not change
Steps:
1.Case base describing problems and solution
2.Match best previous examples
3.Then modify the solution to suit the current problem
4.If solution works add it to the case base
What are the four r’s of CBR cycle?
Retrieved (similar case)
Reuse (Adaption solution)
revise (verify it works)
Retain (learning)
Explain CBR in Insurance
Underwriting Decisions:
– If a new policy proposal is like previous examples, then it can be routinely accepted.
- Claims Handling:
– Refer to similar previous examples - Pricing and Premium Setting:
– Previous examples can be a starting point for calculations - Fraud Detection
- Regulatory compliance
– If a case is similar to a previously compliant one
What is Machine Learning?
Machine learning is an approach to the field of Al Systems trained to recognize patterns within data to acquire knowledge
How does machine learning process differ from traditional programming process?
Traditional: Computer takes in data and programs and produces an output
Machine learning: Takes data and outputs and produces a program
Explain difference between machine learning and statistics
Statistics
– Based on mathematical principles
– Requires data to meet requirements for those techniques
– Prediction less important than understanding
Machine Learning
– Based on what works
– Solutions proposed from training datasets
– Solutions validated by test datasets
– More flexible in data input
– Prediction important, understanding less so
Explain supervised learning
Provide ML algorithm with curated data set
What are the steps to supervised learning?
Classification: Predicting to which discrete class an entity belongs
Regression: Predicting continuous values of an entity’s characteristic.
Forecasting: Estimation of macro (aggregated) variables such as total monthly sales
Attribute Importance: Identifying the variables (attributes) that are the most important for prediction
Explain unsupervised learning?
Uses machine learning algorithms to analyze and cluster unlabeled datasets.
Explain the steps in unsupervised learning
Clustering: Finding natural groupings in the data.
Association models: Analysing “market baskets” (e.g. combinations of the products that are bought together) – Checking for Correlation without explanation
What are issues with machine learning?
- Garbage in/ Garbage out.
– You have the data, but is it quality data?
– Previous processing of data may create bias. - You still need large computing power
– Cost and energy consumption - Results cannot be easily explained
- If circumstances change then your model may be
useless.
What jobs will have increased chances of automation
Higher skilled jobs have less risk while lower skilled jobs have a very high chance of being redundant
Why do we often used combined systems of AI?
Combination of approaches is the best
-Combining human rules with automated rules
-Using machine learning to establish similarity in case based reasoning
Business analytics - focus is not on the techniques but on the business
Insurance uses of GIS
- Existing policies. Show who and where has policies, what is protected, and for which sum.
- Firmographics. This type shows which
businesses are located side by side. e.g. is a bookshop near a fire station or a firework shop. - Historical loss information. Spatial record of all losses for the entire observation period.
- Location risk index. For customers, each location features a specific risk index.
- Natural disasters. fires, floods, tornadoes, earthquakes, etc
Advantages of ML
Identification of trends and patterns
* Largely automatic, with limited human work needed
- Modern technology
– Now large amounts of data that can be used
– Storage and cloud computing are cheap
– Cloud computing offers ML models
