WO2001045355A2

WO2001045355A2 - Method for measuring and analysing audience on communication networks

Info

Publication number: WO2001045355A2
Application number: PCT/FR2000/003541
Authority: WO
Inventors: Hervé LE JOUAN; Ludovic Dubost; Erwan Arzur; Laurent Nicolas; Laurent Azzopardi
Original assignee: Netvalue S.A.
Priority date: 1999-12-14
Filing date: 2000-12-14
Publication date: 2001-06-21
Also published as: EP1342354A2; WO2001045355A3; AU2526301A

Abstract

The invention concerns a method (20) for measuring and analysing activity on at least a packet data communication network (30) [for example a communication network based on the TCP/IP protocol] of a plurality of terminals (10), using means for collecting data and means for transmitting said collected data to a central equipment comprising a database for storing collected data. Said method comprises a step which consists in acquiring and analysing data concerning data exchanged over the communication network by said terminals, said data comprising at least descriptive data of each packet transmitted and received by each terminal, a step which consists in transmitting said collected data to the central equipment subsequent to analysis of the collected data and generating aggregate databases to supply data representing the activity of a panel of equipment items connected to the communication network. The method is characterised in that it comprises an initial step which consists in setting up and configuring operating parameters for a computer application interacting only with the network layer of the computing platform, excluding the application layer processing exchanged data with the network [browser, FTP software, messaging ] on the individual terminals or on a gateway when the terminals do not allow the recording of such an application.

Description

METHOD OF MEASUREMENT AND ANALYSIS OF HEARING ON NETWORKS

COMMUNICATION.

The present invention relates to the field of audience measurement and analysis on open or closed computer networks of the intranet, extranet or Internet type, and more generally on any type of computer network. The invention more particularly makes it possible to collect, store, transmit and analyze the operations related to the use of the computer network on an individual terminal, in particular a computer, but also any equipment equipped with access to a computer network equipped with a physical access or not to said network (wired or wireless). The growing development of the Internet, in particular in the context of commercial relations, and of all means of communication such as electronic mail, data transfer using the FTP protocol, video or HTTP protocol, requires the use of new means of audience measurement and analysis to study user behavior. The conventional means of measuring and analyzing the audience, in particular the methods using questionnaires associated with user panels, do not make it possible to study concretely the behaviors linked to these new means of communication.

PCT patent application WO98 / 26529 describes a measurement and analysis system which records the use of online and offline services via interactive service devices. The system in particular measures and analyzes access to home pages, advertisements and other resources. The means used consist of the use of activity records sorting between the different types of actions performed by the user and then pre-processing of the data received according to their type. US patent 5,675,510 describes an audience measurement and analysis system making it possible to memorize certain actions performed by the user. The patented system measures, analyzes and transmits information related to the use of a computer through an activity log. This log includes the data linked to predetermined events and makes it possible to know the programs and the communication functions launched by the user. This information is then analyzed to determine the user's usage habits.

This method uses activity logs and messages used by the operating system internally to determine which resources are being used. The main drawback is that it depends on the operating system and gives access to a limited amount of information.

The present invention makes it possible to improve the methods for collecting reliable information concerning the use of a network by a terminal. The invention makes it possible to study all the types of communication linked to a computer network such as the Internet, in particular electronic mail, live communications, multimedia data transfers, games and any other standardized or proprietary communication protocol. using packet information transfer following the IP protocol.

The present invention also makes it possible to facilitate the establishment of panels of individual users in order to improve the quality of the audience studies carried out for any type of actor using an open or closed communication network.

The present invention relates to a method for measuring and analyzing activity on at least one network. data packet communication device as described in claim 1.

The method for measuring and analyzing activity on at least one communication network per data packet, of a plurality of terminals, implements means for collecting data and transmitting said collected data to a central device. comprising a database for storing the information collected. Said method comprises a step of acquiring information relating to the data exchanged on the communication network by the terminals, said information comprising at least the descriptive information of each packet transmitted or received by each terminal, a step of transmitting said information collected to the central equipment, and a subsequent step of analyzing the information collected and generating aggregated databases to provide data representative of the activity of a panel of equipment connected to the communication network. The method is characterized in that it comprises an initial step of installation and configuration of the operating parameters of a computer application interacting with the network layer of the operating system only, excluding the application layer of exploitation of the data exchanged with the network [browser, FTP software, messaging, etc.] on the terminals or on a gateway when the terminals do not allow the registration of such an application.

In a variant, the step of analyzing and acquiring information relating to the data exchanged on the communication network is carried out in real time.

Communication on the communication network is preferably based on the IP protocol. In a variant, the initial installation step consists in installing on terminals a inf ormat ic application of analysis and collection of communication protocol data exchanged by the terminal on the network.

In another variant, the initial step of instal lation c ons is to install a computer application for data collection on gateways

[WAP, UMTS, GPRS, I-MODE, ...] for connection between terminals and the network.

The data collected by the terminals or the gateways is transmitted to a server by the communication network itself, and the activity data is stored in the central equipment.

Depending on the variant, it is possible to:

• memorize the information collected in a specific file of the terminal or in a server accessible by said communication network or by another means.

• perform a step of encoding the information collected. • perform a step of compressing the information collected.

When the process for the exploitation of data from terminals does not allow the installation of the measurement computer application [WAP, UMTS, GPRS, I-

MODE, ...], it preferably includes an initial step of registering the user with a server

[http], said server assigning back to said user a unique identifier; and a step of configuring the user's terminal.

The measurement step includes the recording in a random access memory of all the requests sent and of the responses received by the user as well as the time stamp of said requests. The initial recording step consists of saving in a memory of the measurement server [http] the user's identification information, as well as the request for the home page he has chosen, and the result of the application of a hash function of said request. The terminal configuration step can be carried out by sending a short message (SMS, etc.) containing the user's aggregated address.

The method further comprises, preferably a step of specifying a limited population of terminal users forming a panel, said method comprising a step of acquiring information relating to the data exchanged on the communication network by the sub- set of terminals defined only, and a subsequent step of analyzing the information collected to provide data representative of the activity of the subset of equipment connected to the communication network.

The information collected by the server from the terminals includes, in addition to the descriptive information of each packet sent or received by the terminal, an identifier of the user of said terminal, the method comprising a subsequent step of segmenting the data, by comparing each information received by the server with a database containing a data allocation table.

The method may include an additional step consisting in constituting a database specific to a client user of the method, containing at least all the data collected for the latter, the client client having a computer means for interrogating this specific database. .

At least one targeted survey application can also appear on visualization means of said terminal, the content of which is a function of said information relating to the data exchanged on the network.

A history of the responses to said targeted surveys is produced and recorded at said terminal or at the level of the server which sent said application.

The method comprises a step of classifying the different types of activity on a communication network, a step of specifying a limited population of users forming a panel, said method comprising a subsequent step of synthesizing the information analyzed to provide data representative of the activity of the subset of equipment connected to the communication network.

The method may include an additional step of constituting a database specific to a client user of the method, containing at least all of the data collected for the latter, the client user having computer means for interrogating this database. specific.

In a preferred variant, the step of analyzing the data collected consists in processing at least part of the data comprising an identifier of the internet user [ _Internet user id]; the response to the request made [Id _page ]; the start date of consultation of said response by said Internet user [T _pd ]; and the date of end of consultation of the result of the request by said internet user [T _pf ], the identifier of the session [Id _s ], the start of a session [T _ΞD ] corresponding to the start of a connection to the network by said internet user, the end of said connection [T _sf ].

This step makes it possible to determine from said information, a certain number of indices: • a “Gross day point” index

GDP ”corresponding to the percentage of Internet users in the panel consultant of a panel consulting a given page, multiplied by the number of days of connection per internet user over a given period;

• a GssesP “Gross Session Point” index corresponding to the percentage of Internet users multiplied by the number of Web sessions per Internet user;

• a GPP index corresponding to the percentage of Internet users multiplied by the number of separate WEB pages per Internet user; • a GdisP “Gross display Point” index corresponding to the percentage of Internet users multiplied by the number of pages, distinct or not, displayed by the Internet user;

• a GPP / GdisP Report index; • a GdurP “Gross Duration” index

Point ”corresponding to the percentage of Internet users multiplied by the duration in minutes of connection per Internet user;

• a TARGET REACH index corresponding to the number of Internet users of a given target, for a given domain, compared to the number of individuals of the target;

• a TARGET SHARE index defined by the number of Internet users of the target who have visited a domain (site, sector, etc.) relative to the number of global Internet users who have visited the domain (site, sector, etc.);

• a TARGET DISPLAY SHARE index defined by the number of pages displayed by the target compared to the number of pages displayed for all Internet users on the domain (site, sector, ...); • a BANNER DISPLAY SHARE index defined by the number of advertising banners seen by a target on a domain (site, sector, ...) compared to the number of advertising banners seen by all Internet users on the domain (site, sector , ...); • a PAGES WITH BANNER DISPLAY SHARE index defined by the number of pages with advertising seen by the target on the domain (site, sector, ...) in relation to the number of pages with advertising seen by all Internet users on the domain (site, sector, ...);

• a REACH AFFINITY index defined by the proportion of visitors to the domain (site, sector, etc.) within the target group compared to the proportion of the target group among Internet users;

• a DISPLAY AFFINITY index defined by the number of pages viewed by a target on a site, domain or service, based on the total number of pages viewed on the site, the domain;

• a BANNER AFFINITY index defined by the percentage of banner ads seen by a target on a site, domain or service, based on the total number of banner ads seen on the site, the domain;

• a PAGES WITH BANNER AFFINITY index defined by the percentage of pages with advertising seen by a target on a site, domain or service, compared to the total number of pages with advertising seen on the site, the domain;

The method can further include a subsequent step of detecting advertising information by analyzing the data collected by comparison with a reference base and statistical and behavioral exploitation of said detected information.

The present invention also relates to equipment for measuring activity, characterized in that it comprises a measurement server connected to the Internet, said measurement server comprising a database in which the identifiers and the corresponding requests are recorded. to the home pages requested by panelized users, and a calculator ensuring detection and processing of requests generated by users.

The present invention will be better understood on reading an embodiment given below for purely explanatory purposes, with reference to the appended figure.

The method (20) according to the invention is a method for measuring and analyzing activity on a data packet communication network (30), [for example a communication network based on the TCP / IP protocol] a plurality of terminals (10).

These terminals can constitute, for example, in a certain number of personal computers, PC; (11), portable computers (12), pocket organizers (13), fixed telephones with access to the Internet or Webphone (14), personal assistants (15), terminals for accessing the Internet of the television type (16), of mobile telephones with access to the Internet of the WAP, UMTS, GPRS, I-MODE (17) telephones, or the like.

The method (20) according to the invention implements means of real-time analysis, data collection and transmission of said collected data to a central device comprising a database for storing the information collected. Said method comprises a step of acquiring information relating to the data exchanged on the communication network by said terminals, said information comprising at least the descriptive information of each packet transmitted or received by each terminal allowing the analysis of the protocols used, a step of transmitting said collected information to the central equipment, and a subsequent step of analyzing the collected information and generating aggregated databases to provide data representative of the activity of a panel of equipment connected to the network communication. The method is characterized in that it comprises an initial step of installing and configuring the operating parameters of a computer application, said application interacting with the network layer of the operating system only, excluding the application layer. operating data exchanged with the network [browser, FTP software, messaging, etc.] on individual terminals or on a gateway [when the terminals do not allow the registration of such an application].

When connected to the Internet, information transmitted using the IP protocol is intercepted by a specific program. This information is processed independently of the operating system and the software used to connect. Specific means then make it possible to analyze, filter in real time and collect the data exchanged in order to identify information useful for studying the uses of the different protocols used (HTTP, SMTP, POP3, WAP or any other standardized protocol or owner used) on the computer network.

The invention thus makes it possible to intercept this information in the network layer of the operating system. The invention also applies to any terminal connected or not physically to a computer network such as wireless terminals (WAP mobile phone, GPRS, UMTS, I-Mode ...).

In a variant, the initial installation step consists in installing on terminals a computer application for collecting data from the communication protocol exchanged by the terminal over the network.

In another variant, the initial installation step consists in installing a computer data collection application on a gateway [WAP, UMTS, GPRS, I-MODE, ...] of connection between the terminals and the network.

During a connection, the TCP / IP packets are intercepted and memorized in a specific activity log containing any information that may be useful for studying user behavior. This additional information consists, for example, of a time stamp, an identification of the terminal, an identification of the user or of software in action using network resources.

The use of a representative user panel makes it possible to determine the habits of a population and its subsets, using statistical methods of treatment. In order to facilitate the implementation of this method, each panelist is then provided with an application file for the processing of the information collected. This application file allows the interception of data related to the panelist. It can be installed either on its terminal or on a specific server.

It is also possible to intercept the information exchanged by the use of software implemented in a communication on the computer network, Internet for example. Any use of software allowing the implementation of a communication protocol can be studied by this method.

The system according to the invention also uses implementation tools allowing the installation of the measurement and analysis application, means of encoding and compression of the data used for the activity log, means of specific transfer to the activity log as well as user identification means. Data encoding and compression reduce the memory space occupied by activity logs and improve the security of information contained in these logs during the transfer in particular.

The data collected can be transmitted to one or more servers. These data are decoded and decompressed, then undergo specific processing in order to be inserted into a database. These data are then subjected to a statistical processing allowing to extract the useful information and to create analysis reports of the use of the terminals. The transfer of this data can be carried out according to a specific programming, periodically, or according to individual requests.

All the data collected by the technology according to the invention makes it possible to explore both the uses of the Internet or any type of computer network and the audience of the sites visited by the users.

Advantageously, the invention makes it possible to send a polling window whose content depends on the context of the exchange of information in particular applicable to the analysis and study of electronic commerce. The present invention thus makes it possible to obtain relevant information on the exchanges of information (encrypted or not) which have just been carried out, because on the one hand said survey is carried out quickly after the action and on the other hand because that the questions asked depend on the context of the exchange of information.

Installed on the panelized user terminal, the process runs in the background on his computer and preferably starts automatically when it is started. The process is activated when a connection using the IP protocol is launched by the user.

The method monitors connections by intercepting data packets in real time. It records in an activity log all activities based on Internet protocols but none of the data transmitted and / or received (in accordance with the privacy of the panelists). It allows detailed monitoring of the use of Internet protocols such as HTTP, SMTP, POP3; monitoring of duration, volume and soc and other protocols such as ICQ, Audio / Video, FTP, AOL, Infonie, etc.

It also allows the contextual triggering of a survey application whose content adapts to the type of activity performed by the user. This application interrogates the paneled user thanks to the detection of pre-established parameters recorded by the process. The configuration parameters allowing the launching of the survey application consist, for example, of a database of domain names and type of protocols.

This survey application can be run directly on the user's computer or on a server. The survey application is in the form of a pop-up window that appears on the screen to question the user of the terminal. The user responds with a few clicks of the mouse to the questionnaire which was targeted according to the context and the name of the domain detected, for example. The application takes into account the recurring activity of the user thanks to the recording of a history of survey situations and to the personalization of the triggering parameters of the application.

For the study of electronic commerce for example, the application analyzes in real time the active url of the browser and the protocol used to trigger the survey application. This functionality allows to know the behavior of the user, whatever the coding system of the exchanged data. It also allows the identification of users and the analysis of information at the individual level.

All this data is added in real time to the activity log, which is compressed and encoded before any shipment.

It establishes connections with at least one particular server, according to configurable thresholds, in order to load the content of the activity files into databases.

The automatic transfer of activity logs is carried out according to determined thresholds (configurable remotely) from the panelist's computer to one or more specific servers. As soon as the information file is received by the particular server, a recurring periodic process will recover this file, archive it, and load the data contained in a database of the particular server.

This data contained in a database is then subjected to statistical processing enabling it to extract the information necessary for the establishment of summary reports on the use of terminals and behavior. The processing of this data is carried out according to a specific procedure, periodically, or according to individual requests.

All the data collected by the technology according to the invention makes it possible to explore both the uses of the Internet or of any type of computer network and the audience of the main sites.

For the uses of electronic commerce on the Internet or computer networks, the objective is to provide the keys to understanding the behavior of Internet users. The use of the network is analyzed in its different components by categories of Internet users. The method according to the invention provides summaries of their purchasing, reservations, information, etc. procedures. This synthesis makes it possible in particular to optimize recruitment and loyalty plans.

In practice, the terminal user must first register online on the registration website provided for this purpose. During this registration, he is assigned a unique identification number. This number is recorded by the device and will be contained in each log sent in order to allow the identification of the user and the source of the log received.

In the case of wireless terminals in particular, the invention makes it possible, via a gateway, to intercept and store the activity in a log.

According to the particular mode of implementation, the measurement step comprises recording in a random access memory all the accesses generated by the user as well as the time and the date of detection of said accesses.

The measurement method according to the invention consists in interposing a measurement server between the WAP gateway and the servers of the Internet network.

The wireless terminal user must first register online, for example by accessing an html page constituting a registration form on the site of the audience measurement operator from a computer. connected to the Internet. This registration relates to the identity of the user, personal information to qualify him, and the address of the url of the home page. This information is transmitted to the server in the form of a computer message. The registration server assigns a unique identifier to user and saves the results of a hash function on the home page. The aggregated home page is of the form:

[measurement server url address] separator [user id] separator [url of the user's home page], for example http: / / wap. netvalue. com / wapmeter. html? id = XXXX X & url = YYYY

The part before the question mark corresponds to the url address of the measurement server.

The question mark is the first separator.

The part after the question mark corresponds to the identifier assigned by the measurement server.

The & character constitutes a second separator.

The last part corresponds to the home page summarized in an integer resulting from a hash function. This aggregated address is stored in the memory of the user's wireless terminal, in the area corresponding to the address of the home page.

When the terminal user makes his first request from the links on the displayed home page, the measurement server detects all the http links (url) on the page requested by the user and dynamically transforms them into whole number according to a hashing algorithm, assigning to each url a unique number following the model of the aggregated home page above.

The method thus makes it possible to measure the activity of the identified user of the terminal by storing each of the urls generated in a log file. All the data collected by the technology according to the invention makes it possible to explore both the uses of the Internet or of any type of computer network and the audience of the main sites. Advertising is the subject of a specific study within the various processes for processing the data collected. The invention furthermore comprises a step of searching for advertisements, in the various forms that it can use on the communication network, to which the measured users have been exposed.

For the uses of the Internet or computer networks, the objective is to provide the keys to understanding the behavior of Internet users. Network usage (web, email and other uses) is analyzed in its different components (frequency of use, time spent, breakdowns between uses, type of attachment for e-mails, etc.) by categories of Internet users (and in particular by PMG distribution: Small - Medium - Large Internet users). The application of the process offers the possibility of cross-sorting between Small-Medium-Large Internet users for each use of the network, crossing the characteristics of Internet users (age, gender, PCS ...) with all the uses they make of the network. . The method according to the invention provides results in their marketing procedures, in particular in the optimization of recruitment and loyalty plans.

The process allows the analysis of the weight of each sector of activity on the Web (banking, finance, car manufacturers, transport, entertainment ...) and of the journey of Internet users between sectors: page views, time spent, number of sessions .

It also allows the measurement and analysis of the audience of the main sites by two categories of audience indicators. The first category contains the unit indices: coverage rate (number of unique visitors per month), frequency, average number of days of connection, single pages or different views, page display, time spent on a site, total time spent on sites, share of web activity within sessions ...

The second category includes composite indices. These indices synthesize the coverage and the different unit indices Gross Day Point, Gross Session Point, Gross Page Point, Gros Display Point, Gross Duration Point.

These indicators are monitored for sites with a significant audience (by domain names and by sites: grouping of domains), and by sectors

(calculated on the basis of several thousand categorized sites representing more than 80% of Internet traffic).

Exposure frequency indices

• GDP - "Gross Day Point": Percentage of Internet users multiplied by the number of connection days per Internet user over the period.

• GsesP - "Gross Session Point": Percentage of Internet users multiplied by the number of Web sessions per Internet user. Compared to the share of multiple visitors calculated on the number of sessions, this indicator makes it possible to classify the domains.

• GPP - "Gross Page Point": Percentage of Internet users multiplied by the number of distinct Web pages per Internet user.

• GdisP - "Gross Display Point": Percentage of Internet users multiplied by the number of pages (distinct or not) displayed per Internet user. The GPP / GdisP report measures the fact that Internet users have seen the same page several times.

• GdurP - "Gross Duration Point": Percentage of Internet users multiplied by duration (session in minutes) per Internet user. • Reach: Number of visitors for a domain compared to the number of Internet users.

• Target Reach: Number of target visitors for a given domain compared to the number of target individuals.

• Target Share: Number of target individuals who have visited a domain (site, sector, etc.) compared to the total number of Internet users who have visited the domain (site, sector, etc.). • Target Display Share: Number of pages displayed by the target compared to the number of pages displayed by all Internet users on the domain (site, sector, etc.).

• Banner Display Share: Number of advertising banners seen by a target on a domain (site, sector, etc.) compared to the number of advertising banners seen by all Internet users on the domain (site, sector, etc.) .

• Pages with Banner Display Share: Number of pages with advertising seen by the target on a domain

(site, sector, etc.) compared to the number of pages with advertising seen by all Internet users on a domain (site, sector, etc.).

From these indices, different variants of affinity calculation are possible, notably on coverage, categories of Internet users, sectors or domains ...

• Reach Affinity (affinity on audience share or coverage): This indicator measures the fact that the proportion of visitors to the domain within the target group compared to the proportion of the target group among Internet users. This indicator is higher than 100 if the target group's coverage rate is higher than the coverage rate of all Internet users. Number of visitors belonging to the target on a domain

Number of visitors to a Reach Affinity domain X 100 Number of Internet users belonging to the target

Number of Internet users

• Display Affinity (affinity on page views):: This index is calculated from the number of page views by a target on a site, domain or service compared to the total number of page views on the site, domain or service, (target: man, woman, gender, age group, type of internet user ...)

It measures the probability that a page served randomly on the site, group of sites or service reaches a person of the chosen target. It makes it possible to find the sites, group of sites, services on which the purchase of space (for placing advertising banners) is the most effective, that is to say to most effectively reach a percentage of the chosen target. The formula below is given for a domain. We calculate a formula identical to all levels of aggregation.

Number of pages displayed by the target on the αdomain

Numerous countries displayed on the domain for all internet users

Display Affinity x 100

Number of pages displayed by the target on all domains Number of pages displayed by all Internet users s_.r all domains

The area with the highest banner ad affinity is the area in which it is most effective to place advertising to reach the target, in terms of page views. • Banner Affinity (banner ad affinity): This. index is calculated from the percentage of banner ads seen by a target (target: man, woman, gender, age group, type of internet user ...) on a site, domain or service compared to the total number of banner ads viewed on the site, domain or service.

When the Banner Affinity is greater than 100, it indicates that the proportion of banner ads seen by a target on a site is greater than the proportion of banner ads seen by all users on all sites.

The formula below is given for a domain. We calculate an identical formula at all levels of aggregation (site, sector, sub-domain, etc.).

Number of banner ads seen by the target on the domain

Number of banner ads seen by all

Internet user banner on the Affini domain, x 100 ty

Number of banner ads seen by the target on all _, domains

Many advertising banners seen by all Internet users

The area with the highest banner ad affinity index is the area in which it is most effective to place advertising to reach the target, in terms of banner advertising.

• Pages with banner Affinity: This index is calculated from the percentage of pages with advertising seen by a target (target: man, woman, gender, age group, type of Internet user. .) on a site, domain or service related to the total number of pages with advertising seen on the site, the domain. When the 'Pages with banner Affinity' is greater than 100, it indicates that the proportion of pages with banner ads seen by a target on a site is greater than the proportion of banner ad pages seen by all users on all sites.

The formula below is given for a domain. We compute an identical formula at all levels of aggregation (site, sector, sub-domain, etc.).

Number of ads with advertising seen by the target on the domain

Pages Number of pages with advertising seen by all internet users on the Affinity banner domain x 100

Number of advertising pages seen by the target on all domains Number of advertising pages seen by all Internet users on all domains

The area with the highest affinity index on pages with advertising is the area in which it is most effective to place advertising to reach the target, in terms of pages with advertising.

These indices constitute a tool for measuring the effectiveness of the purchase of advertising space on the Internet, and the cornerstone of the media planning offer. They help and guide clients in positioning, strategic planning, advertising sales and purchasing decisions. These indicators have been developed to meet the growing needs of customers to have behavioral data on Internet users in order to differentiate between sites and other entities on the Internet.

The invention makes it possible to offer, by a secure web application, a generic data analysis tool. It offers possibilities for crossing the socio-demographic characteristics of Internet users (age, sex, PCS ...) and behavioral (sites visited, number of emails, etc.) with the audience of the sites. It makes it possible to establish audience reports for sites or domains and for different uses of the Internet or a computer network. The application also makes it possible to identify and view the advertisements disseminated on the Internet and to analyze the socio-demographic profiles of the Internet users who are exposed to them. This information can be delivered for all types of advertising media: banners, buttons before an analysis by sector and product.

These data are interpreted and aggregated across a range of products.

According to a particular implementation variant, the method according to the invention comprises a step of specifying a subset of equipped terminals which can be the subject of a generic measurement and analysis or limited to a particular population. The method according to this variant comprises a step of acquiring information relating to the data exchanged between the communication network and said subset of terminals only, and a subsequent step of analyzing the information collected to provide data representative of the activity of the subset of equipment connected to the communication network.

The invention is described in the foregoing by way of example. It is understood that a person skilled in the art is able to carry out different variants of the invention without going beyond the scope of the patent.

Claims

1 - Method (20) for measuring and analyzing activity on at least one data packet communication network (30), of a plurality of terminals (10), using data collection means and of transmitting said collected data to a central device comprising a database for storing the collected information, said method comprising a step of analysis and acquisition of information relating to the data exchanged on the communication network by said terminals, said information comprising at least the descriptive information of each packet sent or received by each terminal, a step of transmitting said collected information to the central equipment, and a subsequent step of analyzing the collected information and generating aggregated databases for provide data representative of the activity of a panel of equipment connected to the communication network, ca characterized in that the method comprises an initial step of installing and configuring the operating parameters of a computer application interacting with the network layer of the operating system only, excluding the application layer of data exploitation exchanged with the network [browser, FTP software, messaging, etc.] on individual terminals or on a gateway when the terminals do not allow the registration of such an application.

2 - Method for measuring and analyzing activity on at least one communication network according to claim 1 characterized in that the step of analyzing and acquiring information relating to the data exchanged on the communication network is operated in real time. 3 - Method for measuring and analyzing activity on at least one communication network according to claim 1 or claim 2 characterized in that the communication on the communication network is based on the IP protocol.

4 - Method for measuring and analyzing activity on at least one communication network according to any one of claims 1 to 3, characterized in that said initial installation step consists in installing a computer application on terminals analysis and data collection of the communication protocol exchanged by the terminal on the network.

5 - Method for measuring and analyzing activity on at least one communication network according to any one of claims 1 to 3, characterized in that said initial installation step consists in installing a computer application for collecting data on gateways [WAP, UMTS, GPRS, I-MODE, ...] connecting between terminals and the network.

6 - Method for measuring and analyzing activity on at least one communication network according to at least one of the preceding claims, characterized in that the data collected by the terminals or the gateways are transmitted to a server by the network communication itself, and in that the activity data is stored in the central equipment.

7 - Method for measuring and analyzing activity on a communication network according to at least one of claims 1 to 4 characterized in that the information collected is stored in a specific file of the terminal. 8 - Method for measuring and analyzing activity on at least one communication network according to any one of the preceding claims, characterized in that the information collected is stored in a server accessible by said communication network or by another way.

9 - Method for measuring and analyzing activity on at least one communication network according to any one of the preceding claims, characterized in that a step of encoding the information collected is carried out.

10 - Method for measuring and analyzing activity on at least one communication network according to any one of the preceding claims, characterized in that a step of compressing the information collected is carried out.

11 - Method for measuring and analyzing activity on at least one communication network according to any one of claims 5 to 9 characterized in that it comprises an initial step of registering the user with a server [http], said server assigning in return to said user a unique identifier; and a user terminal configuration step.

12 - Method for measuring and analyzing activity on at least one communication network according to claim 11 characterized in that the measuring step comprises recording in a random access memory all the requests sent and the responses received by the user and the timestamp of said requests. 13 - Method for measuring and analyzing activity on at least one communication network according to claim 11 or 12 characterized in that the initial recording step consists of saving in a memory of the measurement server [http] the user's identification information, as well as the request for the home page he has chosen, and the result of the application of a hash function of said request.

14 - Method for measuring and analyzing activity on at least one communication network according to at least one of claims 11 or 12 characterized in that the terminal configuration step is carried out by sending a current message (SMS, ...) containing the user's aggregated address.

15 - Method for measuring and analyzing activity on at least one communication network according to any one of the preceding claims, characterized in that it includes a step of specifying a limited population of terminal users forming a panel, said method comprising a step of acquiring information relating to the data exchanged on the communication network by the subset of terminals defined only, and a later step of analyzing the information collected to provide data representative of the activity of the subset of equipment connected to the communication network.

16 - Method for measuring and analyzing activity on at least one communication network according to any one of the preceding claims, characterized in that the information collected by the server from the terminals comprises, in addition to the descriptive information of each packet sent or received by the terminal, an identifier of the user of said terminal, the method comprising a subsequent step of segmenting the data, by comparing each information received by the server with a database containing a data allocation table.

17 - Method for measuring and analyzing activity on at least one communication network according to any one of the preceding claims, characterized in that it comprises an additional step consisting in constituting a database specific to a client user of the process, containing at least all the data collected for the latter, the client user having an IT means for querying this specific database.

18 - Method for measuring and analyzing activity on at least one communication network according to at least one of the preceding claims, characterized in that at least one targeted survey application appears on means for viewing said terminal, the content is a function of said information relating to the data exchanged on the network.

19 - Method for measuring and analyzing activity on at least one communication network according to claim 18, characterized in that a history of the responses to said targeted surveys is produced [and recorded] at said terminal.

20 - Method for measuring and analyzing activity on at least one communication network according to claim 18, characterized in that a history of the responses to said targeted surveys is produced [and recorded] at the server which has sent said application. 21 - Method for measuring and analyzing activity on at least one communication network according to any one of the preceding claims, characterized in that it includes a step of classification of the different types of activity on a communication network , a step of specifying a limited population of users forming a panel, said method comprising a subsequent step of synthesizing the information analyzed to provide data representative of the activity of the subset of equipment connected to the communication network.

22 - Method for measuring and analyzing activity on at least one communication network according to any one of the preceding claims, characterized in that the step of analyzing the information collected consists in processing at least part of the data comprising an identifier of the user _[ιncernaute Id]; the response to the request made [Id _page ]; the start date of consultation of said response by said Internet user [T _pd ]; and the date of end of consultation of the result of the request by said internet user [T _{p £} ], the identifier of the session [Id _s ], the start of a session [T _sD ] corresponding to the start of a connection to the network by said internet user, the end of said connection [T _sf ].

23 - Method for measuring and analyzing activity on at least one communication network according to claim 22, characterized in that a “Gross day point GDP” index is determined from said information corresponding to the percentage d 'Internet users of the panel consultant of a panel consultant on a specific page, multiplied by the number of days of connection per internet user over a given period. 24 - Method for measuring and analyzing activity on at least one communication network according to claim 22, characterized in that a GssesP “Gross Session Point” index is determined from said information corresponding to the percentage d multiplied by the number of web sessions per user.

25 - Method for measuring and analyzing activity on at least one communication network according to claim 22, characterized in that a GPP index corresponding to the percentage of Internet users multiplied by the is determined from said information number of separate WEB pages per Internet user.

26 - A method of measuring and analyzing activity on at least one communication network according to claim 22, characterized in that a GdisP “Gross display Point” index corresponding to the percentage d is determined from said information. 'Internet users multiplied by the number of pages, distinct or not, displayed by the Internet user.

27 - Method for measuring and analyzing activity on at least one communication network according to claim 22, characterized in that a GPP / GdisP ratio index is determined from said information.

28 - Method for measuring and analyzing activity on at least one communication network according to claim 22, characterized in that a GdurP “Gross Duration Point” index corresponding to the percentage d is determined from said information 'Internet users multiplied by the duration in minutes of connection per Internet user. 29 - Method for measuring and analyzing activity on at least one communication network according to claim 22, characterized in that a TARGET REACH index corresponding to the number of Internet users is determined from said information a given target, for a given domain, in relation to the number of individuals in the target.

30 - Method for measuring and analyzing activity on at least one communication network according to claim 22, characterized in that a TARGET SHARE index defined by the number of Internet users is determined from said information the target who have visited a domain (site, sector, etc.) in relation to the total number of Internet users who have visited the domain (site, sector, etc.).

31 - Method for measuring and analyzing activity on at least one communication network according to claim 22, characterized in that a TARGET DISPLAY SHARE index defined by the number of pages displayed is determined from said information by the target in relation to the number of pages displayed for all Internet users on the domain (site, sector, ...).

32 - Method for measuring and analyzing activity on at least one communication network according to claim 22, characterized in that a BANNER DISPLAY SHARE index defined by the number of advertising banners is determined from said information viewed by a target on a domain (site, sector, ...) compared to the number of advertising banners seen by all Internet users on the domain (site, sector, ...).

33 - Method for measuring and analyzing activity on at least one communication network according to the claim 22, characterized in that a PAGES WITH BANNER DISPLAY SHARE index is defined from said information defined by the number of pages with advertising seen by the target on the domain (site, sector, etc.) relative to the number of pages with advertising seen by all internet users on the domain (site, sector, ...).

34 - Method for measuring and analyzing activity on at least one communication network according to claim 22, characterized in that one determines from said information a REACH AFFINITY index defined by the proportion of visitors to the domain (site, sector, ...) within the target group compared to the proportion of the target group among Internet users.

35 - Method for measuring and analyzing activity on at least one communication network according to claim 22, characterized in that a DISPLAY AFFINITY index defined by the number of pages viewed per is determined from said information a target on a site, a domain or a service, related to the total number of pages viewed on the site, the domain.

36 - Method for measuring and analyzing activity on at least one communication network according to claim 22, characterized in that a BANNER AFFINITY index defined by the percentage of banner ads seen is determined from said information by a target on a site, domain or service, related to the total number of banner ads seen on the site, the domain.

37 - Method for measuring and analyzing activity on at least one communication network according to the claim 22, characterized in that a PAGES WITH BANNER AFFINITY index defined by the percentage of pages with advertising seen by a target on a site, domain or service, based on the total number of pages, is determined from said information with advertising views on the site, domain or service.

38 - Method for measuring and analyzing activity on at least one communication network according to any one of the preceding claims, characterized in that it includes a subsequent step of detecting advertising information by analyzing the data collected by comparison with a reference base and statistical and behavioral analysis of said detected information.

39 - Equipment for activity measurement, characterized in that it comprises a measurement server connected to the Internet, said measurement server comprising a database in which the identifiers and requests corresponding to the requested home pages are recorded by panelized users, and a calculator ensuring the detection and transformation of requests generated by users.