Method and system for harnessing collective knowledge

Abstract

The present invention provides a method of aggregating information, the method including: receiving at least one first prediction; and rating the at least one first prediction based at least in part on at least one actual event, thereby providing at least one prediction rating.

Description

BACKGROUND OF THE INVENTION

FIELD OF THE INVENTION

The present invention is directed to a method and system for presenting aggregated information and, in one embodiment, to a method and system for presenting aggregated information related to stocks.

DISCUSSION OF THE BACKGROUND

Often, individual member/users of a community possess different types of insight and knowledge with respect to any particular subject. These member/users of the community sometimes interact with each other using message boards or other interactive media. While message boards enable member/users to interact and share information with each other, a need has developed for a way to more effectively sort the information shared by the member/users. The present invention addresses this need.

In one non-limiting example, the present invention provides a method, system, and apparatus for aggregating community member/users' knowledge in an efficient manner, independent of the subject matter of the knowledge. In another non-limiting embodiment, the present invention provides a method, system, and apparatus for aggregating community member/users' knowledge with respect to predictions of future events. The present invention also enables community member/users to rate each other's information and predictions about future events, independent of the subject matter of the information and predictions. Another non-limiting embodiment of the present invention enables community member/users to be rated (e.g., by a computer system) based on the accuracy of the information previously presented by each member/user, as well as by the sentiment of the other community member/users. The present invention is especially useful in the field of stock market analysis.

SUMMARY OF THE INVENTION

To achieve the objects of the present invention, a method, system, and apparatus are provided that enable aggregation of community member/users' information, such as opinions, predictions, and other types of evaluations. This information may be related to any type of subject matter, including (but not limited to): financial instrument performance, financial instrument value, market performance, market value, monetary exchange rates, sports teams' performances, professional athletes' performances, college athletes' performances, national and international events, as well as many other types of subject matter apparent to those of skill in the art.

The present invention also provides, as a non-limiting embodiment, a public index database of “subject snapshots” on individual subjects of interest such as financial instruments (e.g., stocks). These subject snapshots may be broken down by various characteristics (e.g., by a particular advisor, topic, technology related to a stock) or other factors known to those of skill in the art. The subject snapshots bring together the most interesting and relevant information by aggregating user input together with publications by professional information providers (such as advisors' publications).

BRIEF DESCRIPTION OF THE DRAWINGS

A more complete appreciation of the invention and many of the attendant advantages thereof will be readily obtained as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, wherein:

FIG. 1 represents a member/user interface according to an embodiment of the present invention.

FIG. 2 represents a member/user interface according to another embodiment of the present invention;

FIG. 3 represents a non-limiting example of a rating screen presented to a member/user;

FIG. 4 represents another non-limiting example of a member/user interface according to the present invention;

FIG. 5 represents a non-limiting illustration of a computer that may be used to implement the method and/or system of the present invention;

FIG. 6 represents another non-limiting example of a member/user interface according to the present invention;

FIG. 7 illustrates another non-limiting example of a member/user interface according to the present invention;

FIG. 8 represents another non-limiting example of a member/user interface according to the present invention;

FIG. 9 represents yet another non-limiting example of a member/user interface according to the present invention;

FIG. 10 represents another non-limiting example of a member/user interface according to the present invention;

FIG. 11 represents another non-limiting illustration of a member/user interface according to the present invention;

FIG. 12 represents another non-limiting example of a member/user interface according to the present invention; and

FIG. 13 represents another non-limiting example of a member/user interface according to the present invention.

DESCRIPTION OF THE EMBODIMENTS

FIG. 1 illustrates a non-limiting example of a member/user interface according to the present invention. This member/user interface may be provided to any individual seeking to participate in the community. A “member/user” refers to an individual accessing the system and/or method of the present invention.

Generally speaking, Internet users enjoy congregating in a community environment. Through the present invention, it is possible for community member/users to act together to combine intelligence. Through this synthesis of community information, the synthesized information may provide more accurate predictions than the experts in a given subject area. The present invention provides, as one non-limiting aspect, a reward for the community member/users who outperform the experts, such as discounts on future services or other rewards known to those of skill in the art.

A further aspect of the present invention enables monetization of the features of the invention. For example, it is possible to charge community member/users a nominal fee to view the top performers. The top performers may be ranked by industry, active stocks, or a member/user's overall success rate. Other factors known to those of skill in the art may be used to select which performers are chosen as the top performers. Additionally, the number of performers chosen may be varied based on the needs of the application.

It is also possible to charge subscribers different rates to access the features of the present invention based on their individual subscription. For example, a subscriber to the majority of the features available through The Motley Fool website may be charged a lower price than a subscriber to fewer portions of The Motley Fool website. It is also possible to provide different levels of access based on how much a member/user is paying for a subscription. For example, a member/user paying a larger fee for a subscription may pay less to view the top performers than a member/user subscribing at a reduced fee.

According to one non-limiting aspect of the present invention, different levels of access to the member/user interface may be provided, depending on a particular individual's level of subscription. For example, if an individual has paid more money for a subscription to be part of the community, that individual may access all features of the interface. By contrast, if an individual has not paid for a subscription to the community, that member/user might be able to access only rudimentary features of the member/user interface (or may not be able to access the member/user interface at all). The levels of access offered by the provider of the member/user interface (hereafter, “the provider”) may be determined according to the provider's needs, and are within the level of ordinary skill in the art.

In the non-limiting example of FIG. 1, a member/user is presented with a particular stock to rate. The member/user may select this stock from a list of stocks supplied by the provider, or may enter his own selection of a stock available on any stock market. Stock markets may include, but are not limited to, the New York Stock Exchange, NASDAQ, the London Stock Exchange, as well as any other type of market known to those of skill in the art. The present invention is also not limited to stock markets, as explained above, and is equally applicable to any situation in which information may be aggregated and/or performance may be assessed.

In the example of FIG. 1, the “Fool Rating” (also referred to as a “Bull/Bear Rating”) is 50%. This Fool Rating may be determined by what the member/users think about a particular prediction by another member/user or by an advisor. The Fool Rating may also reflect how any member/users believe that a particular stock will out perform the market. For example, Fool Ratings are illustrated for a variety of different stocks in FIGS. 2, 6, 7, 9-1 1, and 13. By way of example, a Fool Rating over 50% means that more than half of the member/users believe that the selected stock will outperform the market.

As another non-limiting aspect of the present invention, the Fool Rating may account for a particular member/user's past performance at making accurate predictions. For example, the weight given to an individual's opinion in calculating the Fool Rating may be reduced if that particular member/user frequently makes inaccurate or incorrect predictions. The weight given to an individual's predictions may also be raised when calculating the Fool Rating if the particular member/user's predictions are frequently accurate or correct. The weighting may account for the relative levels of difficulty for each prediction.

The member/user interface of FIG. 1 also provides the stock symbol, the last known stock price, as well as a link to news information related to the selected stock. The member/user interface presents the member/user with an opportunity to rate the selected stock. The member/user may select options such as “outperform,” “underperform,” “match,” as well as other performance indicators known to those of skill in the art. As explained below, the member/user may compare the selected stock's performance to any given number of benchmarks.

The member/user may also provide commentary supporting his selection. This commentary (referred to as a “60 second pitch”) may optionally be used by other member/users of the community who are assessing the rating member/user's likelihood of successful rating or when making their own ratings. Examples of the input field for the 60 second pitch are shown in FIGS. 1, 6, 7, 9-11, and 13. The provider may also collect the member/users' commentary and may provide an aggregated listing of the commentary to member/users. This aggregated listing of commentary may be provided in any format known to those of skill in the art. Non-limiting examples of aggregated commentary are shown in FIGS. 6 and 7. The user may select any of the listed commentaries to view more details. The amount of information provided to the member/user for each commentary may be determined based on the provider's needs. Additionally, the amount of information provided to each member/user may depend on the subscription rate paid by the member/user, the frequency of use of the present invention by the member/user, as well as other factors.

Another aspect of the invention provides a bull/bear sentiment tool that invites any member/user to submit (e.g., via a form) his prediction as to whether an investment (e.g., a stock, mutual fund, option or future) will outperform or underperform a benchmark during a given time period (e.g., 12 months). The benchmark may include a market-wide index or average (e.g., the Dow Jones Industrial Average, the S&P 500, FTSE 100, or the NASDAQ Composite Index) or a sector specific index (e.g., the semiconductor or transportation indices). Alternatively, it is possible to compare an investment or a group of investment to a user- or system-selected group of investments that may be selected based on a variety of factors, such as the related technology, price to earnings ratios, or other factors known to those of skill in the art. (As used herein, the above indices and groups of investments against which a comparison is made shall be referred to collectively herein as “benchmarks.”) It is also possible to predict that a given investment will outperform a first benchmark while simultaneously underperforming relative to a second benchmark. Moreover, each “investment” used in the prediction may actually include more than one different investment or type of investment without departing from the scope of the invention. For example, a user may provide a first prediction that the average return of a particular stock and a mutual fund will be better than the average return of two benchmarks.

Depending on the member/user's level of access, he may be able to view a selection of the most popular picks by other member/users. These picks may be ranked in a number of ways, including total number of picks by member/users, likelihood of outperforming the market based on member/users' picks, likelihood of underperforming the market according to the member/users' picks, as well as by a weighted percentage based on member/users' past successful picks (e.g., a stock picked by a member/user who frequently correctly predicts the stock's performance might be given a higher rating than a stock picked by a member/user who frequently incorrectly predicts that stock's performance). Examples of access to other member/user's picks are illustrated in FIGS. 2, 8, 9, and 13. The member/users may also access additional information related to the picks offered by the provider, at the information provider's discretion. The most popular picks and other information may be provided based on a time restricted manner (e.g., the most popular picks over the last three months).

Another non-limiting example of the present invention provides for a member/user to access other member/users' individual performance information. The individual performance information may be determined and presented by the provider, at the provider's discretion. A member/user may also access a quantity of most recommended posts, as illustrated in FIG. 6. While FIG. 6 shows that the five most recommended posts from the last three months may be selected, the member/user and/or the provider may select different quantities of postings and different time durations, as desired.

As one non-limiting aspect of the present invention, it is possible to use a detailed algorithm to determine the member/user's individual performance information. A basic formula is as follows:

2/3•(score rating)+ 1/3•(percentage right rating)=raw player rating.

The raw player rating may be expressed as a number between 0 and 100, for example, as the official member/user rating displayed to the member/user.

According to this non-limiting example, every member/user may be rated according to two factors: how often the member/user is correct, and by how much the member/user is correct. The score rating represents the number of percentage points by which a member/user is correct or incorrect across every single contest in which the member/user is active. This score rating also includes a member/user's success both past and present. For instance, if a member/user picked three stocks to outperform, and each of these stocks did outperform, and beat the market respectively by 5.06 percentage points, 12.34 percentage points, and 107.94 percentage points, the member/user's score rating would be (+125.34). If the second stock had underperformed by 12.34 points, the member/user's score would be (+100.66). According to this non-limiting example, it is possible for scores to be negative. In short, the score rating represents the total number of percentage points by which a member/user is ahead or behind in the game.

The “percent right” is the percentage of time that the market agrees with the member/user's selection. If a member/user has 38 total contests, and the member/user is correct in 19 of those contests, the member/user's percent right is 50%. The percent right is computed as a simple percentage of correctness over all past and present contests. (Of course, it would be possible to calculate the percent right based on a selected group of contests as well.) According to this non-limiting example, a member/user winning 20 out of 38 contests would have a percent right score of 52.63%.

As one non-limiting example, it is possible to calculate both the percent right and the score rating over the entire universe of member/users. Thus, for example, if exactly 100 members/users were active, each member/user would be listed from highest score to lowest score, regardless of the member/user's actual scores. The highest scoring member/user would receive a rating of 100 and the lowest scoring member/user would receive a rating of 1. Everyone else would be accordingly rated in between these two endpoints. Of course, it is also possible to rate individual member/users within smaller selected groups, as desired by the provider.

In a situation in which the number of member/users is not divisible by 100, it is possible to take the number 100 and divide by the number of players to obtain the player ratings. For example, for 7,394 member/users, the highest score among the 7,394 member/users would be given a 100, the second highest a rating of (100−0.0135), which is 99.9865, and so on. This 0.0135 number is derived from 100 divided by 7,394. Alternatively, the lowest rated member/user would receive a score of .0135, the second lowest member/user a 0.270, etc. The scores may be rounded to the nearest tenth decimal or other factor, as desired by the provider.

In order to obtain the raw player rating, ( 2/3•the member/user's score rating) is added to ( 1/3•the member/user's percent right). In this example, the raw player rating is not the final rating because force ranking all player ratings between 0-100 provides the easiest display for the provider. Other methods of force ranking are possible, depending on the needs of the provider.

To encourage member/user participation, a member/user must have at least seven open active contests, according to a non-limiting aspect of the invention (the number seven is a non-limiting example, and may be changed based on the discretion of the provider). If at any time a member/user falls below this minimum threshold, he will not earn a visible player rating and is considered inactive. Simply put, member/users who do not maintain enough activity may not receive official ratings.

Of course, the provider may still track all of that member/user's data and predictions. While member/users who do not meet the minimum criteria are considered inactive and will not show up on a hot player list or other game page, these member/users are allowed and encouraged to enter more ratings as soon as possible to reach the minimum at any time. Once the member/user has reached the minimum, all of that member/user's past and present contests may be tracked and may affect stock ratings, etc.

As one non-limiting aspect of the present invention, for any given stock, each member/user may only have one contest for that stock. A member/user may not artificially inflate his score or earn minimum active status by playing a given stock more than once at a time. Of course, it is possible for a member/user to end a given contest on a certain stock and to start a second contest using that stock, since non-concurrently running contests do not artificially inflate the member/user's rating.

Each member/user may select a duration for each contest. There is no maximum or minimum duration required. It is also possible for a provider to measure a member/user's response rate using cookies and e-mails, as well as other tracking methods known to those of skill in the art. If the provider determines that the member/user is no longer active, it is possible for the provider to end that member/user's contest at any time after the member/user has been determined to be inactive.

As another non-limiting aspect of the invention, members/users may be prevented from making initial predictions on stocks that are under $1.00 per share. However, if these stocks arrive at such a share price through bad performance of the stock, then the contest may continue. The provider may select certain stocks not to be included in a contest, based on the provider's discretion. The criteria that a provider may use may include stock price and/or volume of stocks traded, as non-limiting examples.

As another non-limiting aspect of the invention, it is possible to end a contest when a stock price hits $0. By ending the contest, it is possible to prevent a member/user who correctly predicted that the stock would underperform to continue to reap improvement in his performance ranking once the stock has reached a price of $0.00 per share.

FIG. 3 illustrates a Microsoft prediction. This attachment may be viewed as the next step after FIGS. 1 or 2. It shows that on 12/22/04 the member/user predicted that in one week (12/29/04) Microsoft will “outperform” the market. As is evident, on the date member/user made that prediction Microsoft's price was $22.96. If a member/user makes a prediction after the market to which the prediction is tied has closed, the opening price of the stock in that market may be used. Alternatively, it is possible to use the closing price of the stock, or an average of the two prices. Other price calculations may also be used, at the discretion of the provider or as requested by the member/user.

The member/user is not required to identify the price of Microsoft at the time of the prediction—the present invention may track the price at the time of prediction relative to the current price without the prediction price being identified to other member/users. The current or latest price is $27.09 in FIG. 2. In this example, if the member/user predicted that Microsoft would outperform the market, the member/user would be winning.

The member/user may also choose the duration of the prediction. In the non-limiting example of FIG. 3, the choices are 1 week, 1 month, 3 months, 6 months, 9 months, 12 months, 24 months, and 36 months. When a user prematurely ends the prediction duration, such as by selecting the “end contest” option in FIG. 7, the premature withdrawal may impact that member/user's overall performance rating, at the discretion of the provider.

It is also possible for a member/user to view information related to other member/users making related predictions. For example, member/users may view all other predictions related to Microsoft, in the non-limiting examples of FIGS. 8, 9, and 13. Member/users may also view their prediction performances relative to the other member/users.

FIG. 12 illustrates an example of performance information that may be provided to a user. In the example of FIG. 12, the member/user using the “screen name” “TMFDiesel” has an aggregate rating of zero. This aggregate rating reflects that TMFDiesel has made only one prediction, which has not yet met its deadline (in the example of FIG. 12, the deadline is 12/20/05, which at the time of the application is in the future). FIG. 12 also provides stock related information for the member/user's reference with respect to the member/user's prediction. FIG. 3 illustrates performance information in an alternative format.

The prediction information for member/users may be sorted and filtered in any way desired, based on the requirements of either the member/user or the information provider. As a non-limiting example, a member/user may view date, ticker, price, prediction, deadline, percent change, the market used, and whether or not the prediction is presently correct for as many other member/users as the information provider will permit (or as many other member/users as the member/user desires). The member/user may sort this information using methods known to those of ordinary skill in the art. Non-limiting examples of the illustration of the information are shown in FIGS. 7, 9, 10, and 13.

A member/user may also access additional information, such as related issues or updates, additional articles, and proprietary information such as the “Foolish 8” criteria. The Foolish 8 criteria relate to revenues, growth rates, net profit margin, daily dollar volume, insider holdings, share price, relative strength, and operating cash flow for a given corporation. The Foolish 8 criteria are based on a combination of business-related and market-related factors that were chosen to highlight eight attributes that investors should look for in small companies. The first four business-level factors set minimum standards for the following things: earnings and sales growth (at least 25% in both cases), net profit margins (at least 7%), operating cash flow (a positive figure), and insider holdings (at least 10% ownership). Three of the four remaining Foolish 8 factors highlight stock market neighborhoods where good small companies are likely found. These criteria relate to stock attributes: stocks with $1 million-$25 million in daily dollar volume, stocks with $7 minimum share prices, and stocks with relative strength ratings of 90 or higher. These requirements are, in large part, out of a company's control. These factors reflect the market feeling, and may not be related to any business-related factors. The final Foolish 8 criterion sets a $500 million limit on a firm's annual sales. The Foolish 8 criteria are applicable to certain stocks, and less applicable to other stocks. The Foolish 8 criteria, as well as other information, may be presented as most recommended posts, examples of which are illustrated in FIGS. 1, 6, 7, 10, and 11.

Subject snapshots, such as the one in FIG. 4, may be provided as an online page. These snapshots (also referred to “Fool's Eye View”) may appear with statistics, in a format similar to performance statistics appearing on a baseball card, for each stock covered by the provider. When a provider selects a new stock, the stock receives a snapshot. Snapshots on stocks not yet covered may also be requested by a member/user accessing the snapshot interface. The member/user may fill out a short submission form that may include information such as the stock's ticker symbol. On a periodic basis, the member/user requests may be reviewed and selected submitted symbols may be chosen to include new snapshots. Thus, for the present invention, an index of snapshots may be developed.

It is also possible to include links to articles, advice, or other information useful to understanding a given stock. As an additional feature, the Fool's Eye View is searchable (e.g., by stock ticker symbol). Through this feature, a member/user may enter a stock ticker symbol or other factor related to the stock to access the service snapshot for that stock. Proprietary information may also be linked to the service snapshot, so that members/users may access the proprietary information as desired (or as provided by the users).

As an additional feature, it is possible to include all available information for a particular stock, advisor, or group of stocks. As a non-limiting example, it is possible to generate a list of the top five most recommended advisors, stocks, or other postings from a particular discussion board.

This invention enables public expression of the community's estimate of stock performance. The user, when making a performance estimate, may include comments as to why the user selected a particular bull/bear evaluation. Additionally, users may be rewarded for making the correct judgments. The present invention may also provide an indication of business strength and price attractiveness for any particular stock or group of stocks. This information may be gleaned from experts in the field, a particular professional advisor, or staff of The Motley Fool, as well as from other industry sources.

According to the present invention, it is possible to provide a member/user with a notification that his prediction has been received or that other events relevant to his prediction have occurred. This notification may include additional information related to the features of the present invention (such as the member/user ranking system, etc.). The notification may be in the form of an email message, a text message, an instant message, or other form known to those in the art.

FIG. 5 is a schematic illustration of a computer system for performing at least one of the functions of FIGS. 1-4 and 6-13. A computer 100 implements the method of the present invention, wherein the computer housing 102 houses a motherboard 104 which contains a CPU 106, memory 108 (e.g., DRAM, ROM, EPROM, EEPROM, SRAM, SDRAM, and Flash RAM), and other optional special purpose logic devices (e.g., ASICs) or configurable logic devices (e.g., GAL and reprogrammable FPGA). The computer 100 also includes plural input devices, (e.g., a keyboard 122 and mouse 124), and a display card 110 for controlling monitor 120. In addition, the computer system 100 further includes a floppy disk drive 114; other removable media devices (e.g., compact disc 119, tape, and removable magneto-optical media (not shown)); and a hard disk 112, or other fixed, high density media drives, connected using an appropriate device bus (e.g., a SCSI bus, an Enhanced IDE bus, or a Ultra DMA bus). Also connected to the same device bus or another device bus, the computer 100 may additionally include a compact disc reader 118, a compact disc reader/writer unit (not shown) or a compact disc jukebox (not shown). Although compact disc 119 is shown in a CD caddy, the compact disc 1 19 can be inserted directly into CD-ROM drives which do not require caddies. In addition, a printer (not shown) may provide hard copies of the features illustrated in FIGS. 1-4.

As stated above, the system includes at least one computer readable medium. Examples of computer readable media are compact discs 119, hard disks 112, floppy disks, tape, magneto-optical disks, PROMs (EPROM, EEPROM, Flash EPROM), DRAM, SRAM, SDRAM, etc. Stored on any one or on a combination of computer readable media, the present invention includes software for controlling both the hardware of the computer 100 and for enabling the computer 100 to interact with a human user. Such software may include, but is not limited to, device drivers, operating systems and user applications, such as development tools. Together, the computer readable media and the software thereon form a computer program product of the present invention for performing at least one of the functions of FIGS. 1-4. The computer code devices of the present invention can be any interpreted or executable code mechanism, including but not limited to scripts, interpreters, dynamic link libraries, Java classes, and complete executable programs.

Obviously, numerous modifications and variations of the present invention are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims, the invention may be practiced otherwise than as specifically described herein.

Claims

1. A method of aggregating information, the method comprising:

receiving at least one first prediction; and

rating the at least one first prediction based at least in part on at least one subsequent event, thereby providing at least one prediction rating.

2. The method according to claim 1, wherein the rating comprises calculating:

2/3•(score rating)+ 1/3•(percentage right rating) to obtain a raw rating,

wherein the score rating represents the number of percentage points by which the at least one first prediction is correct, and

the percentage right rating represents an accuracy of at least one member/user across a plurality of predictions including the at least one first prediction.

3. The method according to claim 2, further comprising adjusting the raw rating to reflect the at least one member/user's performance relative to a second at least one member/user's performance.

4. The method according to claim 3, wherein the second at least one member/user includes a plurality of member/users.

5. The method according to claim 1, further comprising receiving at least one second prediction.

6. The method according to claim 5, further comprising rating the at least one first prediction based at least in part on the at least one second prediction.

7. The method according to claim 5, wherein the at least one second prediction includes a plurality of predictions.

8. The method according to claim 1, wherein the at least one first prediction relates to a stock.

9. The method according to claim 1, further comprising displaying the at least one prediction rating.

10. The method according to claim 1, wherein the at least one prediction rating includes a plurality of prediction ratings.

11. The method according to claim 1, wherein the at least one first prediction is weighted based at least in part on at least one previous prediction.

12. A system for aggregating information, the system comprising:

means for receiving at least one first prediction; and

a controller configured to rate the at least one first prediction based at least in part on at least one subsequent event, thereby providing at least one prediction rating.

13. The system according to claim 12, wherein the controller is configured to rate the at least one first prediction based at in part on the following calculation:

2/3•(score rating)+ 1/3•(percentage right rating) to obtain a raw rating,

wherein the score rating represents the number of percentage points by which the at least one first prediction is correct, and

the percentage right rating represents an accuracy of at least one member/user across a plurality of predictions including the at least one first prediction.

14. The system according to claim 13, wherein the controller is further configured to adjust the raw rating to reflect the at least one member/user's performance relative to a second at least one member/user's performance.

15. The system according to claim 14, wherein the second at least one member/user includes a plurality of member/users.

16. The system according to claim 12, further comprising means for receiving at least one second prediction.

17. The system according to claim 16, wherein the controller is further configured to rate the at least one first prediction based at least in part on the at least one second prediction.

18. The system according to claim 16, wherein the at least one second prediction includes a plurality of predictions.

19. The system according to claim 12, wherein the at least one first prediction relates to a stock.

20. The system according to claim 12, further comprising displaying the at least one prediction rating.

21. The system according to claim 12, wherein the at least one prediction rating includes a plurality of prediction ratings.

22. The system according to claim 12, further comprising means for weighting the at least one first prediction based at least in part on a previous prediction.