WO2019001688A1 - METHOD AND SYSTEM FOR ANALYZING INVESTMENT PROPOSALS - Google Patents

Abstract

The invention relates to a method for analyzing investment proposals, comprising the steps of: transmitting input data (E) relating to an investment proposal from a customer of a user (N) to a user server (S); receive the input data (E) transmitted using the server; evaluating a portion of the input data (E) received relating to the investment proposal as essential data (E1); evaluating a portion of the input data (E) received relating to the user (N) transmitting the input data (E) as volatile data (E2); make a decision on the investment proposal taking into account the input data (E) evaluated; and generating output data (V) relating to the investment proposal; and outputting the output data (V) generated to the user's clients (N). The invention further relates to an investment proposal analysis system comprising client-capable clients (N) and a central server (S), comprising: a device for receiving input data (E); transmitted by the user's (N) customers relating to an investment proposal; a first processing unit (K) for evaluating a portion of the input data (E) received relating to the investment proposal as essential data (E1); a second processing unit (R) for evaluating a portion of the input data (E) received relating to the user (N) transmitting the input data (E) as volatile data (E2); a decision-making unit for determining a decision on the investment proposal taking into account the evaluated input data (E); and an output unit, for outputting output data (V) relating to the investment proposal to the customers of the user (N).

Description

 Method and system for analyzing investment proposals

The present invention relates to a method and a system for the analysis of investment prediction.

From the prior art, the so-called crowdsourcing is known as a method in which an arbitrarily large group of users as a swarm solves a task. Examples include writing the encyclopedia wikipedia.org or finding optimal solutions using peer-to-peer, as well as rating and commenting on questions. On the one hand, it is about accomplishing an exponentially increasing number of factual and volatile input data, as well as stably performing the processing and outputting of data for a swarm of any size. On the other hand, it is about making it possible for a single user in limited time or with limited capacity for information to obtain a complete overview of the available data.

As an example of another kind of Schwarmdienst, however, the so-called crowdfunding can be considered. However, the emergence as well as the decision-making or solution of a problem is carried out very differently in this type of swarm service. Crowdfunding is usually an open collective, meaning that the number of users viewing a campaign, for example, essentially depends on how many users the initiator of a campaign can draw attention to. The level of consent and the number of users supporting a campaign therefore do not allow conclusions to be drawn about the quality of the investment object concerned. Crowdfunding covers only those users who support the campaign. Thus, a campaign or investment of 99% of all users can be regarded as uninteresting and still end successfully, if the remaining percentage of users support the campaign, for example, with a minimum investment. The success of a campaign can thus in extreme cases depend less on the intelligence of the crush than on the number of users. That is a fundamental difference versus crowdsourcing, which typically involves three forms of user response, namely approval, disapproval, and abstention, in decision-making.

When it comes to decision-oriented crowdsourcing, you always start from a closed swarm. While new users can join or leave the swarm, decision-making always depends on whether and how a majority of users behave through consent, disapproval or abstention. If a minimum participation is not achieved, then no decision is made even if a majority of the remaining users signal a consent. Only the achievement of a minimum participation in combination with a majority approval or rejection of a decision leads to a solution. Given these two fundamental differences for this aspect of the two types of crowd services, a technical solution using a crowdsourcing system faces much more complex requirements.

In addition, there is another fundamental difference between crowdsourcing and crowdfunding. Crowdsourcing is a pure peer-to-peer system, in which every user can perform any function offered by the system, whereby the actions of a user are also logically comprehensible. Thus, a user acts as the initiator of a proposal as well as a commentator and voter or co-decision maker of the same. In crowdfunding, on the other hand, proposals are submitted exclusively by third parties outside Schwann. The user has only the comment and election function. This is important insofar as the interests of all participants, namely joint success, are always the same in crowdsourcing. In crowdfunding, on the other hand, the interests of the third party are not completely congruent and may even be the opposite.

Swarm services always follow the idea that a swarm intelligence is formed by any number of users as it is known, for example, from the biology of bees and ants or from research on artificial intelligence. Swarm intelligence means that the decision of the swarm is better than the average individual decision of each individual of the swarm and that the individual user with lower risk in the decision-making process of the ground. However, the necessary communication between users is a barrier here.

A collection, processing and evaluation of the data transport as well as the feedback of knowledge, facts, analyzes and evaluations as well as the ontological quantification by means of artificial intelligence represent a data and transaction quantity which grows exponentially with the number of users in the swarm. At present, however, only linear detection and processing systems exist, the task of which is connected with the mere representation or reproduction of inputs. On the other hand, a system whose task is completed only with the decision (also due to feedback within the communication between users), the ontological quantification of factual input data and the presentation of a user rating by means of volatile data with respect to a problem presented to the swarm does not exist.

Against this background, the object of the present invention is to provide a method for analyzing investment proposals in which continuous and orderly data processing can be ensured by using the crowdsourcing approach.

For the technical solution, the present invention proposes a method for analyzing investment proposals, comprising the method steps: transmission of input data regarding an investment proposal from a client of a user to a server, receipt of the transmitted input data by the server, evaluation of part of the received input data in terms the investment proposal as core data, evaluation of a part of the received input data regarding the user transmitting the input data as volatile data, making a decision regarding the investment proposal taking into account the evaluated input data, generating output data regarding the investment proposal and outputting the generated output data to the clients of the users.

By the inventive evaluation of a part of the received input data with respect to the user transmitting the input data as Volatile data allows each individual user to rate themselves and each other user from the perspective of relevance. For this purpose, the system evaluates, for example, the retention period of a single user on a platform, his level of activity with regard to entering data and with regard to the agreement on an overall goal to be achieved by Schwann and his success rate. The success rate is to be understood as whether the user's previous data provided for previous activities are being taken up by other users, and the solution thus found is for the benefit of the swarm, the assumption being that the swarm users are differently experienced when it comes to the solution of a problem posed to the swarm goes. Also in nature are in a swarm usually specialists who influence the swarm to the benefit of all to a certain extent. So the swarm of bees follows a worker who has found a flower field, because his knowledge of the position of the flower field makes this worker an expert. According to the invention, a coefficient is formed for each user from the volatile data and makes these coefficients available to the user and all other users. The calculation is done dynamically, which means that the coefficient changes with each additional input of the user or the other users, as well as relative, ie that the coefficient is not only the absolute volatile activity of a user, but this always in relation to all others Users. For example, if the average length of stay of all users on a platform decreases, but the length of time a single user stays on the platform remains constant, the user's coefficient of retention increases, although in absolute terms, nothing has changed in this fleeting behavior.

According to a preferred refinement of the method according to the invention, it is provided that information about the evaluation of the core data as primary output data is made available uncompressed to the clients of the users. This makes it possible for a user using the method according to the invention to analyze all information received by the system with regard to an investment proposal itself.

According to the invention, it can further be provided that, before a decision is made, the core data is compressed by means of ontological analysis. In this way, a compression of the input data can be achieved without loss of understanding, since only a reduction of the actual - -

Information is provided. Such a shortening may be necessary, in particular, so that a swarm of any size remains capable of action. While in a swarm of any size the time all users spend swarming to solve the common problem increases, so the time available to each individual user does not increase. Therefore, a system-side analysis of all factual inputs by means of artificial intelligence is advantageous. In such a problem, it is important to identify relevant keywords, which is to provide the system with a base of relevant words that relate to the problem asked. Advantageously, the system is adapted to compensate for human input errors. In this way it can be avoided that many relevant words remain unobserved and thus the delivered result would be adversely affected. According to the invention, all core data of the users are thus analyzed for their opinion or their informational content, for which purpose the language used in its entirety is to be recorded. Phrases or words are understood and evaluated as such. If, for example, the statements "This stock has a good performance", "This stock has no good performance" and "The performance of the stock is good" are given only the recognition of the words "good" and "performance" would be the valuation of second sentence example incorrect, because the user meant, despite the use of the word "good" exactly the opposite. So it's important that input data is understood in complete sentences. Furthermore, a comparison of the understood input against a word dictionary must be made, which words or phrases ascribes a value by means of a fixed scale, where "very good", for example, the value +10, "no matter" the value 0 and "dreadfully" the value - 10 By analyzing the input of a user, the user's opinion can be expressed as a number According to the invention, word directories can be used which are available as publicly available sources of information.

In a further advantageous embodiment of the method according to the invention, it is further provided that, before a decision is made, the core data is compressed by means of quantification. According to the invention, the simplified representation of a user input is understood to be an ontologically quantified value, by which other users, by merely considering a number instead of the time-consuming reading of a comment, the opinion of another user with respect to a Investment proposal received. Thus, when considering any number of input data, a meaningful numerical result is obtained by which a user, for example in combination with a tag cloud, knows what the core of a discussion of all users participating in a problem is and how the collective opinion of all users is or in what proportion users are for or against a particular solution. Since a swarm of any size always produces large amounts of input data, no user would be able to process all the information. However, the user can use the provided quantified core data to make an informed contribution to the decision-making process and to use his available time for problems in which he can do more with his experience. The swarm of users as a whole becomes more efficient and intelligent without suffering any significant loss of information. By such a configuration, a swarm of any size can be kept functional. After all, if the amount of output data is no longer managed by the users, it is impossible for a steadily growing number of users to participate in the solution of problems in the sense of the swarm principle.

Advantageously, it can further be provided that information about the evaluation of the core data is provided as compressed tertiary output data, preferably in the form of a tag cloud, to the clients of the users. Considering such a tag cloud, a single user receiving the tag cloud output data can instantly capture what a discussion about an investment proposal is, without having to read thousands of inputs.

In another embodiment of the method according to the invention, it is provided that information about the evaluation of the volatile data as secondary output data is provided to the server, preferably in the form of a feedback. In this way, the decision making of the swarm can be made more efficient and thereby optimized, since there are system-side information concerning the users transmitting the input data, which can also be taken into account in the process of decision-making regarding the investment proposal. According to the invention, it is further provided that information about the evaluation of the volatile data is provided as volatile data to the clients of the users. This allows a user to obtain information about other users that he or she may consider when assessing investment proposals by another user.

Advantageously, according to the invention, after the transmission of input data with regard to an investment proposal, the investment proposal is filtered. In this case, it can be provided on the system side that suggestions made by the filtering of users are adjusted in such a way that only qualified suggestions are transmitted to other users. Such a filtering may on the one hand be a filtering on the basis of legal requirements and the like, which necessitate the matching of each proposal, or on the other hand a filtering by the operator providing the method according to the invention.

Furthermore, the object according to the invention is achieved by a system for analyzing investment proposals with clients which can be operated by users and a central server, comprising a device for receiving input data transmitted by the clients of the users with regard to an investment proposal, a first processing unit for evaluating part of the received input data regarding the investment proposal as core data, a second processing unit for evaluating a part of the received input data regarding the user transmitting the input data as volatile data, a decision making unit for determining a decision regarding the investment proposal considering the evaluated input data, and an output unit for outputting output data regarding the investment proposal to the clients of the users.

With the system according to the invention, it is possible to build a network-based service for the analysis of investment proposals with any number of users. An essential aspect here is that the complexity of the system from the perspective of the operator or the provider depends neither on the number of users of the system nor on the amount of data to be processed. The system makes it possible in a large number of users to submit one or more investment proposals in real time by means of primary and secondary input data (core data or volatile data) and processes these into primary, secondary and tertiary output data, which in turn are provided to the participating users. Furthermore, the system according to the invention performs an evaluation of the input data in order to derive therefrom primary and secondary and tertiary output data for each of the investment proposals made.

According to an advantageous embodiment, it is provided in the system according to the invention that the output unit is designed for non-compressed output of the evaluated core data as primary output data to the clients of the users.

Furthermore, according to a further embodiment of the system according to the invention, a device for compressing the core data by means of ontological analysis is provided.

Advantageously, a device for compressing the core data by means of quantification is also provided on the system side. By a system-side ontological analysis with subsequent quantification of the core data can thus be an output of the core data not as text, but as a mere number, which expresses, for example, the quality of an investment proposal on a scale of - 10 to +10 as a numerical value. Such a value is faster to capture for other users than, for example, reading non-compressed core data in the form of comments.

According to another preferred embodiment of the system according to the invention, the output unit is provided for outputting information about the evaluation of the core data as compressed tertiary output data, preferably in the form of a tag cloud, to the clients of the users. Advantageously, the system according to the invention comprises a device for receiving information on the evaluation of the volatile data as fed back secondary output data.

Furthermore, the output unit can advantageously be provided for outputting information about the evaluation of the volatile data as volatile data to the clients of the users. In the system according to the invention, a device for filtering the transmitted input data with regard to an investment proposal is provided in a further advantageous embodiment.

According to the invention, it can further be provided that the system is set up in such a way that it first displays to a user the investment proposals with the greatest user participation, for example as part of activities on a website. The higher the user participation, the higher the ranking on the displayed list.

Furthermore, it can also be provided that, on the system side, the degree of user approval or rejection is determined by the ontological quantification of comments by the participating users about the ranking on the displayed list. The investment, which has already been rated positive by Schwärm, is thus brought closer to the user so that he can use the time available to evaluate investment proposals more efficiently.

The system according to the invention can advantageously have means which make it possible to notify each user of new proposals, for example by means of e-mail notification or the like. In accordance with a further advantageous embodiment of the system according to the invention, each individual user can also have all the suggestions available on a platform filtered or sorted according to user-defined categories or criteria.

In a particularly preferred variant of the system according to the invention it is provided that this is designed as a decentralized system. A mixture of a peer-to-peer structure and a server-client structure is implemented so that the system is extended from the server to all participating clients. This means that the clients no longer act as pure terminals, which act only as an input / output interface for a particular user, but are now used as data processing components of the system. For this purpose, certain software modules are set up on the clients for preprocessing of the respective data arising there, which can be implemented, for example, in the form of corresponding JAVA applets on each client. Such a decentralized preprocessing of the input data avoids an exponential increase of data on the Severs side become. This means that the greater part of the data volumes and arithmetic operations no longer accrue to the server, but instead decentralized on the clients. Thus, the workload, scalable per client, can be evenly distributed.

Further details, features and advantages of the invention will be explained in more detail with reference to the embodiments illustrated in the figures of the drawings. Showing:

1 shows a block diagram of the system according to the invention with data streams between multiple clients and a server;

Fig. 2a is a block diagram illustrating the invention

 System taking into account a processing of primary and secondary input or output data

(Proposal phase);

Fig. 2b is a block diagram illustrating the invention

 System taking into account a processing of primary and secondary input or output data

(Comment phase);

Fig. 2c is a block diagram illustrating the invention

 System taking into account a processing of primary and secondary input or output data (selection phase);

Fig. 3 is a block diagram illustrating the invention

 Systems regarding a user rating;

Fig. 4 is a block diagram illustrating the invention

 Systems for ontological analysis;

Fig. 5 is a block diagram illustrating the invention

 System for quantifying factual input data.

1 shows in a block diagram the system X according to the invention for the analysis of investment proposals, in which data streams between several -

Clients C1, C2, C3 and a server S are shown. In this case, each user N1, N2, N3 of the system X is assigned a client C1, C2, C3. In the present case, the corresponding software modules K, Q, O, R for data processing by the system X are located completely on the server S. The clients C1, C2, C3 essentially serve only as an input or output interface to the system X and do not themselves contain any Intelligence as far as the system X is concerned. For example, a normal PC, a smartphone, a PDA or the like can be used as the client C1, C2, C3, as long as an Internet browser is running on this device, with which the Internet site of a provider of a system X can be called up. In this case, the browser window acts as an input / output interface between a user N1, N2, N3 and the system X. The individual clients C1, C2, C3 are connected via the Internet with the server S as a host. In addition, the server S can be connected by means of a unit W with other data sources on the Internet, as additional input data results of bot searches and / or crawlers for certain keywords, listings in search engines or links in user input can be considered. For each investment, a proposal V1, V2, V3 is created in the system X on the server S. For each proposal V1, V2, V3, a certain number of input data E must be continuously recorded by the system X and a specific number of output data A must be output. Each proposal V1, V2, V3 is implemented by a suitably programmed processing unit K, which is designed in each case to generate from the respective input data E the associated output data A in real time. With an increasing number of users N and an increasing number of proposals V or processing units K, very large amounts of data fall at all times as input signals. The input data E must be analyzed and evaluated by the processing units K of the system X in real time, for example, to generate the values for the current status and the current reinking of the proposal V from the respective input data E of a proposal V as output signals. The system X of FIG. 1 is a nonlinear system with feedback.

Looking qualitatively at the required arithmetic operations T and the generated data set D as a function of the connected n users N and v proposals V, theoretically the following dependencies result for a direct implementation of the system X: Τ (Ε · η ν + Α · ν ^Λ 2) and D (E nv + A k ^A 2). - -

This is explained by the fact that in each of the v proposals V the n input data vectors E of the n users N and the v output data vectors A of the feedback are detected (D) and must be processed (T). In the case of a total of v proposals V, this gives as a basic term (E n + A v) v. It should be noted that each input data vector E and each output data vector A itself again consists of a possibly large amount of individual values and that all operations and data in the server computer S must flow together as a hardware unit. At the same time, the system X performs the processing of the volatile data R, which are then deposited for the evaluation of each user N, this time as object N1, N2, N3 on the server S, which is shown in the lower part of FIG.

In Fig. 2a, which shows a block diagram illustrating the system of the present invention considering processing of primary and secondary input data, the input data E (v, e) of each proposal implementing first processing unit K (k) and a second processing unit R (k) supplied. According to the invention, for each proposal V (v), data is processed in parallel in the first processing unit K (k) and the second processing unit R (k) in a so-called suggestion phase. These first processing units K (k) and second processing units R (k) are arranged linearly, whereby the system X automatically consolidates the input data present as core data E1 and volatile data E2 and reproduces them in a form suitable for the users in their entirety. Even the consolidated amount of the output data A1 still grows at least linearly proportional to the number N of the users. For this reason, according to the invention, the core data E1 are compressed as input data by the modules O and Q to such an extent that any number of text for each proposal V is delivered to a tertiary output C as a tag cloud and an associated opinion barometer to all users N. It follows that each individual user can do without (read in dashed lines) to read all the output data A1 and instead only processes the compressed tertiary information C. As part of the evaluation B (v, a) shown below in FIG. 2a, the users N1, N2, N3 themselves become objects on the server, and in the context of the feedback a change of their evaluation takes place in real time. This is done by splitting the input data E into so-called core data E1 and so-called volatile data E2. On the system side, further processing of the core data E1 takes place to the tertiary Issue C on the condensing processing units O, Q and a quantitative representation of the sum of all tertiary output data as a dual percentage, so that an ontological overview of the most frequently mentioned keywords can be given by tag cloud. This allows each user, despite a limited time frame, to capture the aggregate of all the output data A for each proposal V, without having read and evaluated each one of the comments of other users.

FIG. 2b corresponds in content to FIG. 2a and illustrates how the amount of input data E1 and E2 increases exponentially in a comment phase subsequent to the suggestion phase, since a total of 3 ^Λ already exists for three investment proposals V1, V2, V3 made by users N1, N2, N3 2 input data are to be processed as shown on the left side of Fig. 2b. In the output processing, the two data groups, core data E1 and volatile data E2, are logarithmically reduced by the factor n by the system X in real time and without data loss, and A1 (n, a) and B (n, a) data groups, respectively, to the swarm representing another user N (n). However, with a swarm of any size, even this reduction could still be insufficient to keep the swarm ready for action. Therefore, in the comment phase, the further compression of the input data E1 for each proposal V to a tertiary output C is performed both ontologically and quantitatively. As is already known from the proposal phase, this results in the fact that each individual user can refrain from reading all the output data A1 and instead processes only the compressed tertiary information C. Likewise, due to the rating B (v, a), the users N1, N2, N3 themselves become objects on the server and within the scope of the feedback in real time a change of their rating takes place. This is done by splitting the input data E into so-called core data E1 and so-called volatile data E2. In this case, an evaluation by the users N of the system X takes place in the form of a selection W + (positive attitude to an investment proposal), a selection W (negative attitude to an investment proposal) or an abstention H. On the system side, this results in a number a of primary and secondary Output data A1, A2 in the form of a black decision V +, V- generated (overall successful or unsuccessful proposal). In the comment phase shown in Fig. 2b also other input variables can be considered such as various general factors from the Internet such as -

Results of bot searches and / or crawlers for specific keywords, listings in search engines or links to user input, which are supplied by a unit W (not shown). Thus, for example, user-supplied articles on external websites can be ontologically quantified by means of artificial intelligence on the system side, which can reduce the time required for other users without adversely affecting the quality and quantity of the knowledge.

FIG. 2c also corresponds in content to FIGS. 2a and 2b and illustrates how the amount of input data present as core data E1 and volatile data E2 increases exponentially in a selection phase. In the output processing, the two data groups, core data E1 and volatile data E2, are logarithmically reduced by the factor n in real time and without data loss, and A1 (n, a) and B (n, a), respectively, as data groups on the right Swarm N (n) delivered. In contrast to the suggestion and commentary phase, in this selection phase no further compression of the input data E with respect to an ontology or a quantification is necessary, since within the context of the case shown here it is a vote with a binary outcome, that means that as an outcome an investment proposal with "Yes" (V +) or "no" (V-) is evaluated.

The output data of the respective second processing units R (k) are understood here as volatile data. Volatile data provides information derived from a user's behavior on a given proposal V and can be quantified with respect to all of N user's future activities, including present proposal V. These data are considered fleeting because their evaluation makes the overall system better and more accurate, but does not contribute to the core data of a proposal itself. The system shows a high fault tolerance against the failure or also the delayed arrival of the data from the second processing units R (k).

The volatile data is compared with the core data generated in the first processing units K (k). These are necessary for the functioning of the system X and are required by the server S for the final evaluation of the various proposals on the server S. This applies in particular to the selection phase, since the system X in this an analysis of the users N made selection. Due to the fact that the users here have the possibility to participate by consent, refusal and abstention, the system X has to implement the following calculation given by the operator and / or the swarm N to convert the result to each proposal V into a binary result of To convert V + or V-:

• V + = EP * nP> EN * nN AND SUM (nPnN)> m / 100 * n

V- = SUM (nPnN) <m / 100 ^* n

 V- = EP * nP <EN * nN AND SUM (nPnN)> m / 100 * n

Where EP represents the consent, EN the rejection, nP the number of consents, nN the number of rejections and m the minimum coefficient defined by (system) operator or swarm N.

The first and second processing units K (k) and R (k) evaluate the same information in different ways. The first processing unit K (k) evaluates the input data E (v, e) for the corresponding proposal V (v) and the second processing unit R (k) evaluates the input data for the evaluation of the users N. The first and second processing units K (k) and R (k) are basically not fixed or causally linked together.

By way of illustration, the following constructed extreme examples are considered:

Example 1: All users N only pay attention to a proposal V (1) and only look at this proposal V (1), click on the information provided, comment on it, link the proposal V (1) to social media websites and discuss proposal V (1). Then, the first processing unit K (1) reports positive values and the second processing unit R (1) reports that two other proposals V (2), V (3) are ignored. Accordingly, it is conceivable that the proposals V (2) and V (3) fail due to the minimum coefficient m, although all the users N undergo real-time changes of their rating due to the records E2.

Example 2: All users N pay equal attention to all proposals V (1), V (2), V (3), but 90% of users N agree to invest proposal V (1). Then, the first processing unit K (1) returns positive values because there is some attention and purchases. The second However, the processing unit R (1) assesses the approval of the proposal V (1) as a positive rating for the participating users N, eg in the context of the length of stay or actions carried out by the users.

Example 3: All users N only pay attention to suggestion V (1) and click on the information provided, link the proposal V (1) to social media websites, discuss only this proposal V (1) in Internet forums, etc ., but agree without prolonged stay only to the proposal V (2). Then, the first processing unit K (1) reports positive values for the proposal V (2) and the second processing unit R (1), however, evaluates the approval of the proposal V (2) as well as the length of stay and actions / s of the users N Proposal V (1) as a positive rating for the participating users N.

Referring now to Fig. 3, a block diagram illustrating the system X of the present invention is shown in terms of user rating and shows by way of example in detail how the system X performs the ranking (R) of each user N, presently for the proposal V2, by the user N1 was made. In Fig. 3, a detail of Fig. 2a and 2b is shown at the top left to illustrate which processing step it is here. For simplicity's sake, it does not address all available criteria or even combinations of criteria that could be processed to calculate a user's rating R. Only time spent on relevant websites as well as the number of comments and votes in polls are included in the example. The volatile data E2 of the user N1 does not arise in a vacuum but by activities of the user N1. In parallel to the recorded activities of the user N1 just any number of other users Nn are active. The absolute volatile input data E2 of the user N1 (35 minutes, 2 comments, 7/11 votes) must always be seen and evaluated in relation to the average of all volatile input data E2n of all users Nn. Both the record regarding a particular user and the record with respect to all users are therefore subject to constant changes. In this example, it is assumed that dwell time, comment activity and 1/3 turnout are included in the calculation of the user ranking R and are reported as output data A2 to all users Nn. -

The sub-coefficients for the residence time, the comment activity and the turnout result in this example as follows:

• partial coefficient residence time N1 = (residence time N1 /

 Average residence time Nn) x 33.33

 • Subcoefficient comment activity N1 = (comment activity N1 /

 Average comment activity Nn) x 33.33

 • Partial Coefficient Voter turnout N1 = (Voter turnout ratio N1 /

 Average election participation rate Nn) x 33.33

The theoretically perfect ranking of a user N1 is 100. In this example, the user N1 has a ranking R of 60, which is made available to all users as output data A2 in real time. The calculation of the user rating may further include, for example, the suggestion activity, the success or failure of a proposal, or the success or failure of a voting activity of the swarm.

4 shows a block diagram illustrating the system according to the invention with regard to an ontological analysis. The ontological analysis O and the system-side processing thereof are shown for a proposal made by the users N1 and N2 proposal V2. In Fig. 4, a detail of Fig. 2a and 2b is shown at the top left to illustrate which processing step it is here. For simplicity, it is assumed that only two users N1 and N2 make the core data E1 as the primary input data to the proposal V2, in the form of a comment on the proposal V2. The comments given are simply plain text in plain language for the sake of simplicity. In real life, for example, links to other websites with additional texts could also be included as additional information. The ontological database is a product of the operator of the system X and must be set up specifically for the problem of the swarm. For the sake of simplicity, only a few words of the ontological database for an investment discussion will be presented here by way of example. A comparison of comments E1 with the ontological database creates a tag cloud. However, this is not finished at this intermediate stage. Depending on the language used, the system must additionally match the phrases and phrases of the comments with the word dictionary. The sentences "The DAX is heading for an all-time high "and" the DAX is not heading for an all-time high "both use the word" all-time high ", but for the discussion of rising prices this word is only relevant as a quasi-signal for other users Nn if it As is to be understood in the first sentence example, after the comparison with the word dictionary, the first part of the tertiary output data C2 for the proposal V2 is represented in the form of a tag cloud for the users Nn of the swarm.

5 shows a block diagram for illustrating the system according to the invention with regard to a quantification of factual input data, in the present case for a proposal V2 by the users N1 and N2. In FIG. 5, a detail of FIGS. 2a and 2b is shown at the top left, in order to illustrate what processing step it is in this drawing. For the sake of simplicity, it is again assumed that only two users N1 and N2 make the core data E1 as the primary input data for the proposal V2, in the form of a comment on proposal V2. In contrast to the ontological analysis, the system according to the invention in this step deals exclusively with the "opinion" of the user N1 or N2 with respect to their comments on the proposal V2 In this example, the left comment of the input data E1 expresses a very strong positive opinion about the further price course The right-hand commentary shows a rather subdued negative opinion after being compared with the word dictionary, so the left-hand commentary is quantified with +7 and the right-hand commentary with -4 on the system side It would not make sense to show the quantification of each comment for all users Nn, so in this simplified example, quantification is also condensed to a percentage-weighted sentiment barometer, assuming that each comment is the same G Has not important in the representation of the barometer. However, it is also possible that the weighting of each comment, for example, by the rating of the issuing user (comments from users with high rating R have a higher weight, because these users could be considered as "experts") or by the age of the comment (fresher comments are more important because they are more recent). The exemplary embodiments illustrated in the figures of the drawing and the exemplary embodiments explained in connection therewith merely serve to explain the invention and are not restrictive of it.

LIST OF REFERENCE NUMBERS

A output data

A1 Primary output data (with regard to core data, uncompressed)

A2 Secondary output data (for volatile data, uncompensated)

B Secondary output data provided to users A2

C tertiary output data (condensed, tag cloud)

C1. C2, C3 clients

E input data

E1 core data of the input data E

E2 Volatile data of the input data E

K first processing unit (for core data E1)

N1. N2. N3 users

O Device for compressing the core data E1 by means of ontological analysis

Q device for further compression of the core data E1 by means of quantification

R second processing unit (for volatile data E2)

S server

V1, V2, V3 investment proposals œystem

Claims

Claims: 1. Procedure for analyzing investment proposals, including the procedural steps Transmission of input data (E) regarding an investment proposal from a client of a user (N) to a server (S), Receipt of the transmitted input data (E) by the server, Evaluation of a part of the received input data (E) with regard to the investment proposal as core data (E1), Evaluation of a part of the received input data (E) with regard to the user (N) transmitting the input data (E) as volatile data (E2), Carrying out a decision regarding the investment proposal taking into account the evaluated input data (E), and Generation of output data (V) with regard to the investment proposal and output of the generated output data (V) to the clients of the users (N). 2. The method according to claim 1, characterized in that information about the evaluation of the core data (E1) as the primary output data (A1) are provided uncompressed to the clients of the user (N). 3. The method according to claim 1 or claim 2, characterized in that before carrying out a decision making a compression of the core data (E1) by means of ontological analysis (O). 4. The method according to claim 3, characterized in that before carrying out a decision-making further, a compression of the core data (E1) by means of quantification (Q). 5. The method according to claim 3 or claim 4, characterized in that information about the evaluation of the core data (E1) as compressed tertiary output data (C), preferably in the form of a tag cloud, to the clients of the user (N) are provided. 6. The method according to any one of claims 1 to 5, characterized in that information about the evaluation of the volatile data (E2) as a secondary output data (A2) the server (S), preferably in the form of a feedback, are provided. 7. The method according to any one of claims 1 to 6, characterized in that information about the evaluation of the volatile data (E2) as volatile data (B) to the clients of the user (N) are provided. 8. The method according to any one of claims 1 to 7, characterized in that after the transmission of input data (E) with respect to an investment proposal filtering of the investment proposal takes place. 9. system for analyzing investment proposals with users (N) operable clients and a central server (S), comprising a device for receiving from the clients' clients (N) transmitted input data (E) with respect to an investment proposal, a first processing unit ( K) for evaluating part of the received input data (E) with regard to the investment proposal as core data (E1), a second processing unit (R) for evaluating a part of the received input data (E) with respect to the user (N) transmitting the input data (E) volatile data (E2), a decision-making unit for the determination of a decision regarding the investment proposal, taking into account the evaluated input data (E), and an output unit for outputting output data (V) regarding the investment proposal to the clients of the users (N). 10. System according to claim 9, characterized in that the output unit is further provided for non-compressed output of the evaluated core data (E1) as primary output data (A1) to the clients of the users (N). 11. System according to claim 9 or claim 10, characterized in that further comprises means (O) for compressing the core data (E1) by means of ontological analysis is provided. 12. System according to claim 11, characterized in that further comprises a device (Q) for compressing the core data (E1) is provided by means of quantification. 13. System according to claim 11 or claim 12, characterized in that the output unit for outputting information about the evaluation of the core data (E1) as compressed tertiary output data (C), preferably in the form of a tag cloud, to the clients of the users (N) is provided. 14. System according to any one of claims 9 to 13, characterized in that further comprises a device for receiving information about the evaluation of the volatile data (E2) as fed back secondary output data (A2) is provided. 5. System according to any one of claims 9 to 14, characterized in that the output unit for outputting information about the evaluation of the volatile data (E2) is provided as volatile data (B) to the clients of the user (N). 16. System according to any one of claims 9 to 15, characterized in that the server comprises a device for filtering the transmitted input data (E) with respect to an investment proposal.