Personalized Wealth Management through Case-based Recommender Systems

AI*IA 2014 - XIII AI*IA Symposium on Artificial IntelligenceSpecial Track on AI for Society and Economy

Pisa (Italy) - 12.12.2014

Giovanni Semeraro, Cataldo Musto

Personalized Wealth Management through Case-based

Recommender Systems

one minute on the Web

G.Semeraro, C.Musto - Personalized Wealth Management through Case-based Recommender Systems AI*IA 2014 - Special Track on AI for Society and Economy - Pisa (Italy) - 12.12.14

we can handle 126 bits of information we deal with 393 bits of information

ratio: more than 3x(Source: Adrian C.Ott, The 24-hour customer)

(from Matrix)

decision-making is actually challenging

paradox of choice(Barry Schwartz, TED talk “Why more is less”)

(financial) overloadG.Semeraro, C.Musto - Personalized Wealth Management through Case-based Recommender Systems AI*IA 2014 - Special Track on AI for Society and Economy - Pisa (Italy) - 12.12.14

solution: personalizationG.Semeraro, C.Musto - Personalized Wealth Management through Case-based Recommender Systems AI*IA 2014 - Special Track on AI for Society and Economy - Pisa (Italy) - 12.12.14

to adapt asset portfolios

on the ground of personal user profile and needs

Insight:

SolutionRecommender Systems

Recommender Systems

Relevant items (movies, news, books, etc.) are suggested to the user according to her preferences.

definitionRecommender Systems have the goal of guiding the

users in a personalized way to interesting

or useful objects in a large space of possible options.

Burke, 2002 (*)(*) Robin D. Burke: Hybrid Recommender Systems: Survey and Experiments. UMUAI, volume 12, issue 4, 331-370 (2002)

does it fit our scenario?“we are leaving the age of information, we are entering the age of recommendation”

(C.Anderson, The Long Tail. Wired. October 2004)

Recommender Systems

“[...] The technology is used by shopping websites such as Amazon, which receives about 35 percent of its revenue via product recommendations. It is also used by coupon sites like Groupon; by travel sites to suggest flights, hotels, and rental cars; by social-networking sites such as LinkedIn; by video sites like Netflix to recommend movies and TV shows, and by music, news, and food sites to suggest songs, news stories, and restaurants, respectively. Even financial-services firms recently began using recommender systems to provide alerts for investors about key market events in which they might be interested”

(N.Leavitt, “A technology that comes highly recommended” - http://tinyurl.com/d5y5hyl)

Recommender Systemssuccess stories

“People who bought…”on Amazon

“Discover”on Spotify

(unexpected) success stories

Recommender SystemsRecommender Systems

recommending financial products is a complex task

flocking

Too many users could be moved towards the same suggestions

flocking

consequence: price manipulation (as in trader forums)

poor knowledge

Features describing both assets classes and private investors are

poorly meaningful

poor knowledge

Solution

Case-based Recommender SystemsG.Semeraro, C.Musto - Personalized Wealth Management through Case-based Recommender Systems AI*IA 2014 - Special Track on AI for Society and Economy - Pisa (Italy) - 12.12.14

case-based RSs• Inspired by case-based reasoning

• Similar problems solved in the past are used as knowledge base

• Reasoning by analogy

• The recommendation process relies on the retrieval and the adaptation of the solutions adopted to solve similar cases

....butwhat do we actually mean with ‘case’ ?

case base

• A case is a the formalization of a previously solved problem

• In our setting

• Description of a user

• Description of a portfolio

• An evaluation of the proposed solution

case-baseexample

user solution evaluation

case-baseexample

User Features Risk Profile: LowFinancial Experience: HighFinancial Situation: Very HighInvestment Goals: MediumTemporal Goals: Medium

case-baseexample

Obbligazionario Euro Bot 30%

Obbligazionario High Yield 10%

Obbligazionario Globale 22%

Azionario Europa 23%

Azionario Paesi Emergenti 7%

Flessibili Bassa Volatilità 8%

case-baseexample

monthly rate (e.g.)

+0.22%

case-based RSssolving cycle

case-based reasoning for personalized wealth management

scenario

“Scrooge McDuck wants to get richer. He decided to invest some of his savings and he asked for help to a

financial advisor”

step 1 user modeling

scenario

Which features may describe

Scrooge McDuck?

step 1

scenario

User Features Risk Profile: Low

Financial Experience: HighFinancial Situation: Very HighInvestment Goals: MediumTemporal Goals: Medium

step 1

scenario

User Features Risk Profile: Low

Financial Experience: HighFinancial Situation: Very HighInvestment Goals: MediumTemporal Goals: Medium

MiFID-based

step 1

in a classical pipeline, the target user

would have received a “model” portfolio tailored on her profile

in a pipeline fostered by a recommender system, the financial advisor can analyze the portfolios proposed to similar users

to tailor the proposal

step 2 retrieval of similar users

given a case base, it is necessary to

define a similarity measure to compute how similar two cases are

given a case base, it is necessary to

define a similarity measure to compute how similar two cases are

vector space representation

cosine similarity to get the most similar usersG.Semeraro, C.Musto - Personalized Wealth Management through Case-based Recommender Systems AI*IA 2014 - Special Track on AI for Society and Economy - Pisa (Italy) - 12.12.14

scenario

case base

step 2

scenariostep 2

similarity score

scenariostep 2

neighborhood(helpful cases)

scenario

Obbligazionario Euro Bot 30%Obbligazionario High Yield 15%Obbligazionario Globale 15%

Azionario Europa 20%Azionario Paesi Emergenti 12%

Obbligazionario Euro Bot 30%Obbligazionario High Yield 10%Obbligazionario Globale 22%

Azionario Europa 23%Azionario Paesi Emergenti 7%

step 2

solutions proposed to the neighbors are labeled as candidate solutions

step 3 revise of the solution

in real-world scenarios, the case base

contains many helpful cases

in real-world scenarios, the case base

contains many helpful cases

it is necessary to introduce strategies to filter and rank the cases

step 3

We defined two ranking strategies

• Diversification

• Financial Confidence Value (FCV)

revise

diversification

insight: filtering out too similar solutions

revise

revisediversification

identification of the best subset of similar cases which maximize the relative diversity

reviseObbligazionario Euro Bot 30%

diversification

reviseObbligazionario Euro Bot 30%

Flessibili Bassa Volatilità 8% XXdiversification

reviseFCV

• Simple insight

• We know the historical yield for each of the assets class in the portfolio

• FCV ranks first the solutions composed by a combination of asset classes close to the optimal one (according to previous yield)

reviseFCV

(Generated yield) (Drift Factor)Total yield is the product of the

yield generated by each asset

class with the its percentage in the

portfolio

Ratio between the yield

generated by the asset classes in the portfolio and its complement

Obbligazionario Euro Bot --- 30%

Azionario Europa +++ 20%

reviseFCV

step 4 review

financial advisor and private investor

can further discuss the portfolio

review

Original Discussed Gap

Obbligazionario Euro Bot 30% 30%

Obbligazionario High Yield 12.5% 10% -2.5%

Obbligazionario Globale 18.5% 20% +1.5%

Azionario Europa 21.5% 24% +2.5%Azionario Paesi

Emergenti 9.5% 8% -1.5%Flessibili Bassa

Volatilità 8% 8%

interactive personalization

step 5 retain

an evaluation score is finally assigned to the proposed solution

yield, e.g.

retain

good solutions are stored in the case base and exploited for future recommendations

case base

(new) case base

evaluation

evaluationwhat is the average yield of

recommended portfolios?

evaluationwhat is the average yield of

recommended portfolios?

can recommender systems suggest

better investment portfolios than human advisors?

experiment 1revise strategies (leave-one-out evaluation)

best performing configuration provides 0,28% monthly yield

neighbors

1 5 10

0,250,240,22

0,270,28

0,220,2

0,150,13

0,160,180,18

Basic Diversification FCV FCV + Div

experiment 2comparison to baselines (leave-one-out evaluation)

recsys better than humans!

neighbors

1 5 10

0,270,28

0,220,20,20,2

0,170,170,17

Human Collaborative FCV

experiment 3ex-post evaluation (6 months, with real data)

FCV and Diversification is the best one

neighbors

1 5 10

0,060,060,060,040,04

0,110,12

0,090,1

0,060,08

Basic FCV FCV + Div Collaborative Human

• Personalized Wealth Management

• Application of case-based reasoning

• Geometrical similarity measure to identify the most similar previously solved cases

• Introduction of diversification and re-ranking techniques

• More than 3% yield for year

• Experiments shows that recommended portfolios overcome the real ones for almost all the users

• Working Demo!

questions?Giovanni Semeraro

giovanni.semeraro@uniba.it

Cataldo Musto cataldo.musto@uniba.it

Personalized Wealth Management through Case-based Recommender Systems

economy pisa italy

hybrid recommender systems

definitionrecommender

ia symposium

adrian c

personalized way

giovanni semeraro

economypisa italy

Technology

Research on Personalized Recommender System for Tourism...

Introduction to a Recommender System for Apache OFBiz ·...

A Recommender Model for the Personalized Adaptive CHUNK ...

Design and Implementation of a Customer Personalized...

Recommender Systems for Mass Customization of Financial...

RECOMMENDER TV A Personalized TV Guide System Compliant...

Recommender System Application Developments: A Survey · PDF...

Profile Injection Attack Detection in Recommender System ·...

User Profiling based on Folksonomy Information in Web 2.0....

Blending Human Computing and Recommender Systems for...

Mealionaire: A context-aware and ontology-based mobile...

Dynamic generation of personalized hybrid recommender...

Hybrid Recommender System Architecture for Personalized...

Yum-Me: A Personalized Nutrient-Based Meal Recommender...

Personalized recommender system on whom to follow in Twitter

Personalized Recommender System Using Entropy Based...