Wiki**3 - User contributions [en]

Sistema de Recomendação (projecto)

2025-07-02T11:04:01Z

Root:

Este projecto destina-.se a realizar um sistema de recomendação baseado em filtros colaborativos.

== Collaborative Filtering Recommendation System ==

Automatic recommendation algorithms are important in electronic commerce for matching users with products that may be of interest to them. A simple but effective way of achieving this match is called collaborative filtering. There are several versions of collaborative filtering recommendation systems, depending on the exact nature of the recommendation goal. In general, these algorithms start from a matrix of users by products and produce recommendations or rating predictions, by matching users (or products), using similarity measures between users (or products).

In our scenario, the users-products matrix encodes the associations between users and products they acquire. Furthermore, users-products pairs are unique, i.e., a user may acquire a product more than once, but there will be only one record connecting that user with that product. Although this binary representation has some limitations (it does not reflect product preferences, for instance), it makes processing simpler.

When computing suggestions for a given user, the algorithm first finds users that are similar to the one under consideration and, then, from the most popular products acquired by the most similar users, selects the products to recommend. User similarity is computed in terms of the products that they acquire and can be computed in various ways. A popular way is to consider the cosine distance, measured between the vectors formed by the products acquired by each user: each vector position corresponds to a product (considering all known products acquired by all users) and is, in our scenario, either 0 (the user has not acquired the product yet), or 1 (the user has already acquired the product).

<center>[[image:sim-cosine.png|500px]]</center>

In this formula ''u1'' and ''u2'' are two users, ''vecu1'' and ''vecu2'' are the corresponding product vectors, ''Pu1'' and ''Pu2'' are the sets of products each user has acquired. ||·|| is the vector norm operator and |·| is the set cardinality operator.

== Running the Program ==

The program takes property db to specify the file containing the users-products database and a command line argument for specifying the user to provide recommendations for.

The output consists of a list of 10 products (one per line), ordered by their fit to the user or alphabetically if tied.

You may assume that the database does not contain errors and that there are no mistakes when invoking the program. As an example, if the database file is '''last.fm.100k.txt''' and the user to get recommendations for is '''user_000577''', the command to invoke the recommender is (App is the class containing the main method):

java -Ddb=last.fm.100k.txt App user_000577

== Solution ==

Class '''Store''' implements the store and the recommendation algorithm.

{{CollapsedCode|Store.java|
<source lang="java">
package scollafil;

import java.io.BufferedReader;
import java.io.FileReader;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Collections;
import java.util.HashMap;
import java.util.HashSet;
import java.util.List;
import java.util.Map;
import java.util.Set;

/**
* This class represents a store (with products and clients).
*
* The store keeps track of what was bought by whom and vice-versa.
*
* @author David Martins de Matos
*/
@SuppressWarnings("nls")
public class Store {
/**
* This constant controls the size of the top-N list for candidate selection.
* It is used both to select the closest clients and the most interesting
* products.
*/
private final int TOP_N = 10;

/** Each entry lists buyers of a given product. */
private Map<String, HashSet<String>> _clientsByProduct = new HashMap<>();

/** Each entry lists products bought by a given client. */
private Map<String, HashSet<String>> _productsByClient = new HashMap<>();

/**
* @param db
*/

public Store(String db) {
try {
BufferedReader reader = new BufferedReader(new FileReader(db));
String line = "";
while ((line = reader.readLine()) != null) {
String[] fields = line.split("\t");
String client = fields[0], product = fields[1];

HashSet<String> cbps = _clientsByProduct.get(product);
if (cbps == null) {
cbps = new HashSet<>();
_clientsByProduct.put(product, cbps);
}
cbps.add(client);

HashSet<String> pbcs = _productsByClient.get(client);
if (pbcs == null) {
pbcs = new HashSet<>();
_productsByClient.put(client, pbcs);
}
pbcs.add(product);
}
reader.close();
} catch (IOException e) {
System.err.println("Error occurred while opening " + db + ": " + e);
System.exit(-1);
}
}

/**
* @param client
* @return recommendation list (a list of product names)
*/
public List<String> getRecommendationsFor(String client) {
ArrayList<Distance> distances = new ArrayList<>();
for (String other : _productsByClient.keySet()) {
if (other.equals(client))
continue;
distances.add(cosine(client, other));
}
Collections.sort(distances);

System.out.println("Closest " + TOP_N + " to " + client + ":");
HashMap<String, Integer> counts = new HashMap<>();

for (int i = 0; i < distances.size() && i < TOP_N; i++) {
Distance d = distances.get(i);
if (d == null)
break;
String target = d.getTarget(); // similar client
System.out.printf("[%d]@%.2g%% = %s\n", i, d.getDistance() * 100, target);

for (String p : _productsByClient.get(target)) {
Integer cc = counts.get(p);
counts.put(p, cc == null ? 0 : cc + 1);
}
}

Set<String> ks = counts.keySet();
ks.removeAll(_productsByClient.get(client));

ArrayList<Distance> bestProducts = new ArrayList<>();
for (String k : ks)
bestProducts.add(new Distance(client, k, counts.get(k)));
Collections.sort(bestProducts);

ArrayList<String> recomm = new ArrayList<>();
for (int i = 0; i < bestProducts.size() && i < TOP_N; i++) {
Distance d = bestProducts.get(i);
if (d == null)
break;
recomm.add(d.getTarget());
System.out.printf("[%d]@%g = %s\n", i, d.getDistance(), d.getTarget());
}

return recomm;
}

/**
* @param c1
* @param c2
* @return "cosine" of two sets
*/
public Distance cosine(String c1, String c2) {
Set<String> v1 = _productsByClient.get(c1), v2 = _productsByClient.get(c2);
Set<String> intersection = new HashSet<>(v1);
intersection.retainAll(v2);
double d = intersection.size() / Math.sqrt(v1.size()) / Math.sqrt(v2.size());
return new Distance(c1, c2, d);
}

}
</source>
}}

Class '''Distance''' is used to encode distances between users and products.

{{CollapsedCode|Distance.java|
<source lang="java">
package scollafil;

/**
* Distance between two points: source and target.
*
* @author David Martins de Matos
*/
class Distance implements Comparable<Distance> {
/** Source. */
private String _source;

/** Target. */
private String _target;

/** Distance. */
private double _distance;

/**
* @param source
* @param target
* @param distance
*/
public Distance(String source, String target, double distance) {
_source = source;
_target = target;
_distance = distance;
}

/**
* Natural order is according to distance, except when distances are same. In
* this case, the order is alphabetical (target).
*
* @see java.lang.Comparable#compareTo(java.lang.Object)
*/
@Override
public int compareTo(Distance d) {
return _distance > d._distance ? -1
: (_distance < d._distance ? 1 : _target.compareTo(d._target));
}

/** @return source */
public String getSource() {
return _source;
}

/** @return target */
public String getTarget() {
return _target;
}

/** @return distance */
public double getDistance() {
return _distance;
}

}
</source>
}}

Class '''App''' runs the program: it asks the user for a name and provides recommendations, based on the history of the store.

{{CollapsedCode|App.java|
<source lang="java">
package scollafil;

import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStreamReader;

/**
* This class implements a simple collaborative filtering recommender.
*
* The program uses property "db" to read a shopping database. After
* initialization, client names are requested and a product list is produced for
* each one. Only the top-3 items are presented.
*
* @author David Martins de Matos
*/
public class App {
/**
* @param args
* command line arguments
*/
@SuppressWarnings("nls")
public static void main(String[] args) {
String db = System.getProperty("db");
if (db == null) {
System.err.println("Please, provide a shopping database!");
return;
}
Store store = new Store(db);
BufferedReader in = new BufferedReader(new InputStreamReader(System.in));
String client = "";

try {
do {
System.out.print("Client: ");
if ((client = in.readLine()) == null || client.equals(""))
break;
System.out.println(store.getRecommendationsFor(client));
} while (true);
} catch (IOException e) {
System.err.println("Error while reading standard input!");
e.printStackTrace();
System.exit(1);
}
}
}
</source>
}}

Projectos

2023-11-23T15:47:58Z

Root: Created page with "<Inesc-id what='card' id='09889818-fc9e-4b29-b0d8-04028590c6b4'/> A informação seguinte é mantida pelo INESC-ID Lisboa.<br/> Actualmente, está apenas disponível em inglês. <Inesc-id what='projects' id='09889818-fc9e-4b29-b0d8-04028590c6b4'/> category:Investigação en:Projects"

2023-11-23T11:58:19Z

Root:

A informação seguinte é mantida pelo INESC-ID Lisboa. Actualmente, está apenas disponível em inglês.

<inesc-id what='advising' id='09889818-fc9e-4b29-b0d8-04028590c6b4'></Inesc-id>

[[category:Orientação Científica]]
[[en:Scientific Advising]]

Apresentação

2023-08-09T09:12:09Z

Root:

Template:Infobox

2023-08-07T18:42:59Z

Root:

Template:Infobox

2023-08-07T18:40:09Z

Root:

{| style="margin-left: 10px; margin-bottom: 10px; width: 200px; font-size: 95%; border: 1px solid; background: #dee2ff;" cellpadding="0" align="right"
! align="center" style="padding: 0px;" | <div style="border-style: solid; border-width: 3px; width: 200px;">[[Image:tipo-passe-{{{username}}}.png|200px|center|{{{name}}}]]</div>
|}

<div style="align: right; margin-left: 10px; margin-bottom: 10px; border-style: solid; border-width: 3px; width: 200px;"><img src="[[media:tipo-passe-{{{username}}}.png]]" /></div>

2020-02-21T14:31:38Z

Root:

{{infobox|name=David Martins de Matos|username=david|contact=david.matos
|phone=+351-213-100-305|fax=+351-213-145-843
}}
Sou professor associado do [http://www.dei.tecnico.ulisboa.pt/ Departamento de Engenharia Informática] do [http://www.tecnico.ulisboa.pt/ Instituto Superior Técnico]. Sou investigador do [http://www.l2f.inesc-id.pt/ Laboratório de Tecnologias de Língua Humana] do [http://www.inesc-id.pt INESC-ID] .

Os meus interesses científicos incluem gestão de recursos linguísticos; arquitecturas de software para engenharia da língua natural; geração automática de língua natural e construção automática de resumos; interfaces de língua natural; e processamento de língua escrita em geral.

== CV Oficial (resumido) ==

''Prof. David Martins de Matos''

David Martins de Matos licenciou-se em Engenharia Electrotécnica e de Computadores em 1990, no Instituto Superior Técnico (IST), Lisboa. Recebeu os grau de Mestre em Engenharia Electrotécnica e de Computadores em 1995 (IST), sob o tema programação com objectos em sistemas distibuídos. Obteve o grau de Doutor em Engenaria Informática e de Computadores em 2005 (IST), sobre o tema da geração automática de língua natural. É docente no IST desde 1993 -- actualmente como Professor Auxiliar --, tendo leccionado as disciplinas de programação em lógica e funcional, programação com objectos, algoritmos e estruturas de dados, compiladores, arquitectura de sistemas computacionais, sistemas distribuídos, computação gráfica e processamento de língua natural. Foi investigador no INESC desde 1988, nos Grupos de Sistemas Distribuídos, Serviços e Sistemas Telemáticos e Engenharia de Software. Em 1998 passou a fazer parte do Centro de Sistemas Telemáticos e Computacionais (IST), onde permaneceu até 2001, quando passou a fazer parte do recém-criado Laboratório de Sistemas de Língua Falada (L2F). No passado, participou em vários projectos Europeus e Nacionais, assim como em projectos do sector privado (banca e indústria de telecomunicações), tanto nacionais como internacionais. É membro sénior da ACM (SIGMM, SIGIR) e do IEEE (Signal Processing Society, Computer Society, Robotics and Automation Society). É também membro da Ordem dos Engenheiros.

[[en:Introduction]]

2020-02-16T22:21:27Z

Root:

/* CSS placed here will be applied to all skins */
/* Custom Foreground menubar colors */
.top-bar,
.top-bar-section ul,
.top-bar-section ul li.active > a,
.top-bar-section li a:not(.button),
.top-bar-section .has-form {
background: #282727;
color: #dddddd;
}

.top-bar-section > ul > .divider,
.top-bar-section > ul > [role="separator"] {
border-color: #282727;
}

/* Darker color for the hover over items */
.top-bar-section ul li.hover > a,
.top-bar-section ul li ul.dropdown li.hover > a{
background: #282727;
}

/* Slightly darker color for the dropdowns */
.top-bar-section ul li ul.dropdown li > a {
background: #282727;
}
.top-bar .button {
background: #414141;
border-top-color: #555555;
border-right-color: #414141;
border-bottom-color: #555555;
border-left-color: #414141;
}

body {
font-family: "Lato", Helvetica, sans-serif;
font-size: 13px;
line-height: 1.25;
color: #697060;
/*background-color: #f7f7f7;*/
background-color: #ffffff;
margin: 0px;
}

body #mw-content-text pre {
margin-bottom: 1em;
}

body span.mw-headline {
font-weight: normal;
}

body span.mw-collapsible-toggle.mw-collapsible-toggle-collapsed {

}
body span.mw-collapsible-toggle.mw-collapsible-toggle-expanded {
}

body h1 { font-size: 1.85em; }
body h2 { font-size: 1.75em; }
.ns-subject h3, .ns-special h3, .page-Main_Page h3, body h3 { font-size: 1.4em; }
body h4 { font-size: 1em; }
body h5 { font-size: 1em; }
body h6 { font-size: 1em; }

body h1,
body h2,
body h3,
body h4,
body h5,
body h6 {
font-family: "Candara", "Lato", sans-serif;
font-weight: normal;
color: #697060;
margin-top: 0.8em;
}
body h2.title { font-size: 2em; }
body .title {
font-family: "Candara", "Lato", sans-serif;
font-weight: normal;
color: #697060;
margin-top: 0em;
}
body div#firstHeading {
font-family: "Candara", "Lato", sans-serif;
font-weight: normal;
color: #697060;
margin-top: 0em;
}

body div#toctitle h2 {
font-size: initial;
margin-top: 0.8em;
display: inline;
}

body #catlinks {
margin-top: 1.5em;
margin-bottom: 0em;
border-style: none;
border-width: 0px;
}

body div#catlinks {
margin-top: 1.5em;
margin-bottom: 0em;
border-style: none;
border-width: 0px;
}

body #catlinks a.label {

}
body #mw-normal-catlinks {
margin-top: 0em;
margin-bottom: 0em;
border-style: none;
border-width: 0px;
}
body .group {
border-style: none;
border-width: 0px;
}
body #mw-normal-catlinks a.label {

}

body pre {
font-family: "Droid Sans Mono", monospace;
font-size: 0.85em;
line-height: 1.25em;
background-color: #fcfcfc;
padding: 8px;
border-style: solid;
border-width: 1px;
border-color: #dddddd;
}

@media all {
body h3#tagline {
display: none;
font-weight: normal;
}
}

@media print {
body h3#tagline { display: block; font-weight: normal; }
a[href]::after { display: none; }
div.printfooter { display: none; }
}

MediaWiki:Common.css

2020-02-16T22:19:26Z

Root:

MediaWiki:Common.css

2020-02-16T22:18:16Z

Root: