I am having a hard time with the question below. I am not sure if I got it correct, but either way, I need some help futher understanding it if anyone has time to explain, please do.
Design L1 and L2 distance functions to assess the similarity of bank customers. Each customer is characterized by the following attributes:
− Age (customer’s age, which is a real number with the maximum age is 90 years and minimum age 15 years)
− Cr (“credit rating”) which is ordinal attribute with values ‘very good’, ‘good, ‘medium’, ‘poor’, and ‘very poor’.
− Av_bal (avg account balance, which is a real number with mean 7000, standard deviation is 4000)
Using the L1 distance function computes the distance between the following 2 customers: c1 = (55, good, 7000) and c2 = (25, poor, 1000). [15 points]
Using the L2 distance function computes the distance between the above mentioned 2 customers
Using the L2 distance function computes the distance between the above mentioned 2 customers.
Answer with L1
d(c1,c2) = (c1.cr-c2.cr)/4 +(c1.avg.bal –c2.avg.bal/4000)* (c1.age-mean.age/std.age)-( c2.age-mean.age/std.age)
The question as is, leaves some room for interpretation. Mainly because similarity is not specified exactly. I will try to explain what the standard approach would be.
Usually, before you start, you want to normalize values such that they are rougly in the same range. Otherwise, your similarity will be dominated by the feature with the largest variance.
If you have no information about the distribution but just the range of the values you want to try to nomalize them to [0,1]. For your example this means
norm_age = (age-15)/(90-15)
For nominal values you want to find a mapping to ordinal values if you want to use Lp-Norms. Note: this is not always possible (e.g., colors cannot intuitively be mapped to ordinal values). In you case you can transform the credit rating like this
cr = {0 if ‘very good’, 1 if ‘good, 2 if ‘medium’, 3 if ‘poor’, 4 if ‘very poor’}
afterwards you can do the same normalization as for age
norm_cr = cr/4
Lastly, for normally distributed values you usually perform standardization by subtracting the mean and dividing by the standard deviation.
norm_av_bal = (av_bal-7000)/4000
Now that you have normalized your values, you can go ahead and define the distance functions:
L1(c1, c2) = |c1.norm_age - c2.norm_age| + |c1.norm_cr - c2.norm_cr |
+ |c1.norm_av_bal - c2.norm_av_bal|
and
L2(c1, c2) = sqrt((c1.norm_age - c2.norm_age)2 + (c1.norm_cr -
c2.norm_cr)2 + (c1.norm_av_bal -
c2.norm_av_bal)2)
I am trying to find a solution in which a given resource (eg. budget) will be best distributed to different options which yields different results on the resource provided.
Let's say I have N = 1200 and some functions. (a, b, c, d are some unknown variables)
f1(x) = a * x
f2(x) = b * x^c
f3(x) = a*x + b*x^2 + c*x^3
f4(x) = d^x
f5(x) = log x^d
...
And also, let's say there n number of these functions that yield different results based on its input x, where x = 0 or x >= m, where m is a constant.
Although I am not able to find exact formula for the given functions, I am able to find the output. This means that I can do:
X = f1(N1) + f2(N2) + f3(N3) + ... + fn(Nn) where (N1 + ... Nn) = N as many times as there are ways of distributing N into n numbers, and find a specific case where X is the greatest.
How would I actually go about finding the best distribution of N with the least computation power, using whatever libraries currently available?
If you are happy with allocations constrained to be whole numbers then there is a dynamic programming solution of cost O(Nn) - so you can increase accuracy by scaling if you want, but this will increase cpu time.
For each i=1 to n maintain an array where element j gives the maximum yield using only the first i functions giving them a total allowance of j.
For i=1 this is simply the result of f1().
For i=k+1 consider when working out the result for j consider each possible way of splitting j units between f_{k+1}() and the table that tells you the best return from a distribution among the first k functions - so you can calculate the table for i=k+1 using the table created for k.
At the end you get the best possible return for n functions and N resources. It makes it easier to find out what that best answer is if you maintain of a set of arrays telling the best way to distribute k units among the first i functions, for all possible values of i and k. Then you can look up the best allocation for f100(), subtract off the value this allocated to f100() from N, look up the best allocation for f99() given the resulting resources, and carry on like this until you have worked out the best allocations for all f().
As an example suppose f1(x) = 2x, f2(x) = x^2 and f3(x) = 3 if x>0 and 0 otherwise. Suppose we have 3 units of resource.
The first table is just f1(x) which is 0, 2, 4, 6 for 0,1,2,3 units.
The second table is the best you can do using f1(x) and f2(x) for 0,1,2,3 units and is 0, 2, 4, 9, switching from f1 to f2 at x=2.
The third table is 0, 3, 5, 9. I can get 3 and 5 by using 1 unit for f3() and the rest for the best solution in the second table. 9 is simply the best solution in the second table - there is no better solution using 3 resources that gives any of them to f(3)
So 9 is the best answer here. One way to work out how to get there is to keep the tables around and recalculate that answer. 9 comes from f3(0) + 9 from the second table so all 3 units are available to f2() + f1(). The second table 9 comes from f2(3) so there are no units left for f(1) and we get f1(0) + f2(3) + f3(0).
When you are working the resources to use at stage i=k+1 you have a table form i=k that tells you exactly the result to expect from the resources you have left over after you have decided to use some at stage i=k+1. The best distribution does not become incorrect because that stage i=k you have worked out the result for the best distribution given every possible number of remaining resources.
I want to analyse some data in order to program a pricing algorithm.
Following dates are available:
I need a function/correlationfactor of the three variables/dimension which show the change of the Median (price) while the three dimensions (pers_capacity, amount of bedrooms, amount of bathrooms) grow.
e.g. Y(#pers_capacity,bedroom,bathroom) = ..
note:
- in the screenshot below are not all the data available (just a part of it)
- median => price per night
- yellow => #bathroom
e.g. For 2 persons, 2 bedrooms and 1 bathroom is the median price 187$ per night
Do you have some ideas how I can calculate the correlation/equation (f(..)=...) in order to get a reliable factor?
Kind regards
One typical approach would be formulating this as a linear model. Given three variables x, y and z which explain your observed values v, you assume v ≈ ax + by + cz + d and try to find a, b, c and d which match this as closely as possible, minimizing the squared error. This is called a linear least squares approximation. You can also refer to this Math SE post for one example of a specific linear least squares approximation.
If your your dataset is sufficiently large, you may consider more complicated formulas. Things like
v ≈
a1x2 +
a2y2 +
a3z2 +
a4xy +
a5xz +
a6yz +
a7x +
a8y +
a9z +
a10
The above is non-linear in the variables but still linear in the coefficients ai so it's still a linear least squares problem.
Or you could apply transformations to your variables, e.g.
v ≈
a1x +
a2y +
a3z +
a4exp(x) +
a5exp(y) +
a6exp(z) +
a7
Looking at the residual errors (i.e. difference between predicted and observed values) in any of these may indicate terms worth adding.
Personally I'd try all this in R, since computing linear models is just one line in that language, and visualizing data is fairly easy as well.
For my work, I need some kind of algorithm with the following input and output:
Input: a set of dates (from the past). Output: a set of weights - one weight per one given date (the sum of all weights = 1).
The basic idea is that the closest date to today's date should receive the highest weight, the second closest date will get the second highest weight, and so on...
Any ideas?
Thanks in advance!
First, for each date in your input set assign the amount of time between the date and today.
For example: the following date set {today, tomorrow, yesterday, a week from today} becomes {0, 1, 1, 7}. Formally: val[i] = abs(today - date[i]).
Second, inverse the values in such a way that their relative weights are reversed. The simplest way of doing so would be: val[i] = 1/val[i].
Other suggestions:
val[i] = 1/val[i]^2
val[i] = 1/sqrt(val[i])
val[i] = 1/log(val[i])
The hardest and most important part is deciding how to inverse the values. Think, what should be the nature of the weights? (do you want noticeable differences between two far away dates, or maybe two far away dates should have pretty equal weights? Do you want a date which is very close to today have an extremely bigger weight or a reasonably bigger weight?).
Note that you should come up with an inverting procedure where you cannot divide by zero. In the example above, dividing by val[i] results in division by zero. One method to avoid division by zero is called smoothing. The most trivial way to "smooth" your data is using the add-one smoothing where you just add one to each value (so today becomes 1, tomorrow becomes 2, next week becomes 8, etc).
Now the easiest part is to normalize the values so that they'll sum up to one.
sum = val[1] + val[2] + ... + val[n]
weight[i] = val[i]/sum for each i
Sort dates and remove dups
Assign values (maybe starting from the farthest date in steps of 10 or whatever you need - these value can be arbitrary, they just reflect order and distance)
Normalize weights to add up to 1
Executable pseudocode (tweakable):
#!/usr/bin/env python
import random, pprint
from operator import itemgetter
# for simplicity's sake dates are integers here ...
pivot_date = 1000
past_dates = set(random.sample(range(1, pivot_date), 5))
weights, stepping = [], 10
for date in sorted(past_dates):
weights.append( (date, stepping) )
stepping += 10
sum_of_steppings = sum([ itemgetter(1)(x) for x in weights ])
normalized = [ (d, (w / float(sum_of_steppings)) ) for d, w in weights ]
pprint.pprint(normalized)
# Example output
# The 'date' closest to 1000 (here: 889) has the highest weight,
# 703 the second highest, and so forth ...
# [(151, 0.06666666666666667),
# (425, 0.13333333333333333),
# (571, 0.2),
# (703, 0.26666666666666666),
# (889, 0.3333333333333333)]
How to weight: just compute the difference of all dates and the current date
x(i) = abs(date(i) - current_date)
you can then use different expression to assign weights:
w(i) = 1/x(i)
w(i) = exp(-x(i))
w(i) = exp(-x(i)^2))
use gaussian distribution - more complicated, do not recommend
Then use normalized weights: w(i)/sum(w(i)) so that the sum is 1.
(Note that the exponential func is always used by statisticians in survival analysis)
The first thing that comes to my mind to to use a geometric series:
http://en.wikipedia.org/wiki/Geometric_series
(1/2)+(1/4)+(1/8)+(1/16)+(1/32)+(1/64)+(1/128)+(1/256)..... sums to one.
Yesterday would be 1/2
2 days ago would be 1/4
and so on
Is is the index for the i-th date.
Assign weights equal to to Ni / D.
D0 is the first date.
Ni is the difference in days between the i-th date and the first date D0.
D is the normalization factor
converts dates to yyyymmddhhmiss format (24 hours), add all these values and the total, divide by the total time, and sort by this value.
declare #data table
(
Date bigint,
Weight float
)
declare #sumTotal decimal(18,2)
insert into #Data (Date)
select top 100
replace(replace(replace(convert(varchar,Datetime,20),'-',''),':',''),' ','')
from Dates
select #sumTotal=sum(Date)
from #Data
update #Data set
Weight=Date/#sumTotal
select * from #Data order by 2 desc
I'm trying to sort a bunch of products by customer ratings using a 5 star system. The site I'm setting this up for does not have a lot of ratings and continue to add new products so it will usually have a few products with a low number of ratings.
I tried using average star rating but that algorithm fails when there is a small number of ratings.
Example a product that has 3x 5 star ratings would show up better than a product that has 100x 5 star ratings and 2x 2 star ratings.
Shouldn't the second product show up higher because it is statistically more trustworthy because of the larger number of ratings?
Prior to 2015, the Internet Movie Database (IMDb) publicly listed the formula used to rank their Top 250 movies list. To quote:
The formula for calculating the Top Rated 250 Titles gives a true Bayesian estimate:
weighted rating (WR) = (v ÷ (v+m)) × R + (m ÷ (v+m)) × C
where:
R = average for the movie (mean)
v = number of votes for the movie
m = minimum votes required to be listed in the Top 250 (currently 25000)
C = the mean vote across the whole report (currently 7.0)
For the Top 250, only votes from regular voters are considered.
It's not so hard to understand. The formula is:
rating = (v / (v + m)) * R +
(m / (v + m)) * C;
Which can be mathematically simplified to:
rating = (R * v + C * m) / (v + m);
The variables are:
R – The item's own rating. R is the average of the item's votes. (For example, if an item has no votes, its R is 0. If someone gives it 5 stars, R becomes 5. If someone else gives it 1 star, R becomes 3, the average of [1, 5]. And so on.)
C – The average item's rating. Find the R of every single item in the database, including the current one, and take the average of them; that is C. (Suppose there are 4 items in the database, and their ratings are [2, 3, 5, 5]. C is 3.75, the average of those numbers.)
v – The number of votes for an item. (To given another example, if 5 people have cast votes on an item, v is 5.)
m – The tuneable parameter. The amount of "smoothing" applied to the rating is based on the number of votes (v) in relation to m. Adjust m until the results satisfy you. And don't misinterpret IMDb's description of m as "minimum votes required to be listed" – this system is perfectly capable of ranking items with less votes than m.
All the formula does is: add m imaginary votes, each with a value of C, before calculating the average. In the beginning, when there isn't enough data (i.e. the number of votes is dramatically less than m), this causes the blanks to be filled in with average data. However, as votes accumulates, eventually the imaginary votes will be drowned out by real ones.
In this system, votes don't cause the rating to fluctuate wildly. Instead, they merely perturb it a bit in some direction.
When there are zero votes, only imaginary votes exist, and all of them are C. Thus, each item begins with a rating of C.
See also:
A demo. Click "Solve".
Another explanation of IMDb's system.
An explanation of a similar Bayesian star-rating system.
Evan Miller shows a Bayesian approach to ranking 5-star ratings:
where
nk is the number of k-star ratings,
sk is the "worth" (in points) of k stars,
N is the total number of votes
K is the maximum number of stars (e.g. K=5, in a 5-star rating system)
z_alpha/2 is the 1 - alpha/2 quantile of a normal distribution. If you want 95% confidence (based on the Bayesian posterior distribution) that the actual sort criterion is at least as big as the computed sort criterion, choose z_alpha/2 = 1.65.
In Python, the sorting criterion can be calculated with
def starsort(ns):
"""
http://www.evanmiller.org/ranking-items-with-star-ratings.html
"""
N = sum(ns)
K = len(ns)
s = list(range(K,0,-1))
s2 = [sk**2 for sk in s]
z = 1.65
def f(s, ns):
N = sum(ns)
K = len(ns)
return sum(sk*(nk+1) for sk, nk in zip(s,ns)) / (N+K)
fsns = f(s, ns)
return fsns - z*math.sqrt((f(s2, ns)- fsns**2)/(N+K+1))
For example, if an item has 60 five-stars, 80 four-stars, 75 three-stars, 20 two-stars and 25 one-stars, then its overall star rating would be about 3.4:
x = (60, 80, 75, 20, 25)
starsort(x)
# 3.3686975120774694
and you can sort a list of 5-star ratings with
sorted([(60, 80, 75, 20, 25), (10,0,0,0,0), (5,0,0,0,0)], key=starsort, reverse=True)
# [(10, 0, 0, 0, 0), (60, 80, 75, 20, 25), (5, 0, 0, 0, 0)]
This shows the effect that more ratings can have upon the overall star value.
You'll find that this formula tends to give an overall rating which is a bit
lower than the overall rating reported by sites such as Amazon, Ebay or Wal-mart
particularly when there are few votes (say, less than 300). This reflects the
higher uncertainy that comes with fewer votes. As the number of votes increases
(into the thousands) all overall these rating formulas should tend to the
(weighted) average rating.
Since the formula only depends on the frequency distribution of 5-star ratings
for the item itself, it is easy to combine reviews from multiple sources (or,
update the overall rating in light of new votes) by simply adding the frequency
distributions together.
Unlike the IMDb formula, this formula does not depend on the average score
across all items, nor an artificial minimum number of votes cutoff value.
Moreover, this formula makes use of the full frequency distribution -- not just
the average number of stars and the number of votes. And it makes sense that it
should since an item with ten 5-stars and ten 1-stars should be treated as
having more uncertainty than (and therefore not rated as highly as) an item with
twenty 3-star ratings:
In [78]: starsort((10,0,0,0,10))
Out[78]: 2.386028063783418
In [79]: starsort((0,0,20,0,0))
Out[79]: 2.795342687927806
The IMDb formula does not take this into account.
See this page for a good analysis of star-based rating systems, and this one for a good analysis of upvote-/downvote- based systems.
For up and down voting you want to estimate the probability that, given the ratings you have, the "real" score (if you had infinite ratings) is greater than some quantity (like, say, the similar number for some other item you're sorting against).
See the second article for the answer, but the conclusion is you want to use the Wilson confidence. The article gives the equation and sample Ruby code (easily translated to another language).
Well, depending on how complex you want to make it, you could have ratings additionally be weighted based on how many ratings the person has made, and what those ratings are. If the person has only made one rating, it could be a shill rating, and might count for less. Or if the person has rated many things in category a, but few in category b, and has an average rating of 1.3 out of 5 stars, it sounds like category a may be artificially weighed down by the low average score of this user, and should be adjusted.
But enough of making it complex. Let’s make it simple.
Assuming we’re working with just two values, ReviewCount and AverageRating, for a particular item, it would make sense to me to look ReviewCount as essentially being the “reliability” value. But we don’t just want to bring scores down for low ReviewCount items: a single one-star rating is probably as unreliable as a single 5 star rating. So what we want to do is probably average towards the middle: 3.
So, basically, I’m thinking of an equation something like X * AverageRating + Y * 3 = the-rating-we-want. In order to make this value come out right we need X+Y to equal 1. Also we need X to increase in value as ReviewCount increases...with a review count of 0, x should be 0 (giving us an equation of “3”), and with an infinite review count X should be 1 (which makes the equation = AverageRating).
So what are X and Y equations? For the X equation want the dependent variable to asymptotically approach 1 as the independent variable approaches infinity. A good set of equations is something like:
Y = 1/(factor^RatingCount)
and (utilizing the fact that X must be equal to 1-Y)
X = 1 – (1/(factor^RatingCount)
Then we can adjust "factor" to fit the range that we're looking for.
I used this simple C# program to try a few factors:
// We can adjust this factor to adjust our curve.
double factor = 1.5;
// Here's some sample data
double RatingAverage1 = 5;
double RatingCount1 = 1;
double RatingAverage2 = 4.5;
double RatingCount2 = 5;
double RatingAverage3 = 3.5;
double RatingCount3 = 50000; // 50000 is not infinite, but it's probably plenty to closely simulate it.
// Do the calculations
double modfactor = Math.Pow(factor, RatingCount1);
double modRating1 = (3 / modfactor)
+ (RatingAverage1 * (1 - 1 / modfactor));
double modfactor2 = Math.Pow(factor, RatingCount2);
double modRating2 = (3 / modfactor2)
+ (RatingAverage2 * (1 - 1 / modfactor2));
double modfactor3 = Math.Pow(factor, RatingCount3);
double modRating3 = (3 / modfactor3)
+ (RatingAverage3 * (1 - 1 / modfactor3));
Console.WriteLine(String.Format("RatingAverage: {0}, RatingCount: {1}, Adjusted Rating: {2:0.00}",
RatingAverage1, RatingCount1, modRating1));
Console.WriteLine(String.Format("RatingAverage: {0}, RatingCount: {1}, Adjusted Rating: {2:0.00}",
RatingAverage2, RatingCount2, modRating2));
Console.WriteLine(String.Format("RatingAverage: {0}, RatingCount: {1}, Adjusted Rating: {2:0.00}",
RatingAverage3, RatingCount3, modRating3));
// Hold up for the user to read the data.
Console.ReadLine();
So you don’t bother copying it in, it gives this output:
RatingAverage: 5, RatingCount: 1, Adjusted Rating: 3.67
RatingAverage: 4.5, RatingCount: 5, Adjusted Rating: 4.30
RatingAverage: 3.5, RatingCount: 50000, Adjusted Rating: 3.50
Something like that? You could obviously adjust the "factor" value as needed to get the kind of weighting you want.
You could sort by median instead of arithmetic mean. In this case both examples have a median of 5, so both would have the same weight in a sorting algorithm.
You could use a mode to the same effect, but median is probably a better idea.
If you want to assign additional weight to the product with 100 5-star ratings, you'll probably want to go with some kind of weighted mode, assigning more weight to ratings with the same median, but with more overall votes.
If you just need a fast and cheap solution that will mostly work without using a lot of computation here's one option (assuming a 1-5 rating scale)
SELECT Products.id, Products.title, avg(Ratings.score), etc
FROM
Products INNER JOIN Ratings ON Products.id=Ratings.product_id
GROUP BY
Products.id, Products.title
ORDER BY (SUM(Ratings.score)+25.0)/(COUNT(Ratings.id)+20.0) DESC, COUNT(Ratings.id) DESC
By adding in 25 and dividing by the total ratings + 20 you're basically adding 10 worst scores and 10 best scores to the total ratings and then sorting accordingly.
This does have known issues. For example, it unfairly rewards low-scoring products with few ratings (as this graph demonstrates, products with an average score of 1 and just one rating score a 1.2 while products with an average score of 1 and 1k+ ratings score closer to 1.05). You could also argue it unfairly punishes high-quality products with few ratings.
This chart shows what happens for all 5 ratings over 1-1000 ratings:
http://www.wolframalpha.com/input/?i=Plot3D%5B%2825%2Bxy%29/%2820%2Bx%29%2C%7Bx%2C1%2C1000%7D%2C%7By%2C0%2C6%7D%5D
You can see the dip upwards at the very bottom ratings, but overall it's a fair ranking, I think. You can also look at it this way:
http://www.wolframalpha.com/input/?i=Plot3D%5B6-%28%2825%2Bxy%29/%2820%2Bx%29%29%2C%7Bx%2C1%2C1000%7D%2C%7By%2C0%2C6%7D%5D
If you drop a marble on most places in this graph, it will automatically roll towards products with both higher scores and higher ratings.
Obviously, the low number of ratings puts this problem at a statistical handicap. Never the less...
A key element to improving the quality of an aggregate rating is to "rate the rater", i.e. to keep tabs of the ratings each particular "rater" has supplied (relative to others). This allows weighing their votes during the aggregation process.
Another solution, more of a cope out, is to supply the end-users with a count (or a range indication thereof) of votes for the underlying item.
One option is something like Microsoft's TrueSkill system, where the score is given by mean - 3*stddev, where the constants can be tweaked.
After look for a while, I choose the Bayesian system.
If someone is using Ruby, here a gem for it:
https://github.com/wbotelhos/rating
I'd highly recommend the book Programming Collective Intelligence by Toby Segaran (OReilly) ISBN 978-0-596-52932-1 which discusses how to extract meaningful data from crowd behaviour. The examples are in Python, but its easy enough to convert.