JP6056061B2 - Information posting system - Google Patents

Description

  The present invention relates to an information posting system for sequentially posting information about a provider who provides a product or service on one screen of an information terminal.

  With the development of electronic commerce in recent years, an increasing number of websites display providers that provide products and services in descending order based on some standard on one screen. In addition, as the number of such websites increases, a technique for adjusting which descending order a provider is displayed on one screen has been proposed (see, for example, Patent Documents 1 to 4).

JP 2010-134485 A JP 2005-537591 A JP 2009-217418 A JP 2004-213377 A

  However, in Patent Documents 1 to 4 described above, the ranking is determined based on each algorithm, and the providers are uniformly displayed on one screen in descending order based on the ranking. For this reason, a provider with a higher ranking is always displayed from the top of the screen, and a provider with a lower ranking is not scrolled or the page is switched a considerable number of times by the user. It was a situation that never caught the eye of the user.

  For this reason, there is a problem that a provider with a lower ranking has a large disparity with a provider with a higher ranking, and a provider with a lower ranking is registered in the posting service. Since this itself does not make sense, there are cases where the service is canceled one after another. For this reason, there was a possibility that the posting service itself was discarded.

  In addition to the evaluation from users who have made past transactions with the provider and the estimated amount of the provider, the ranking is set comprehensively from the positional relationship between the location information of the user who wants to provide products or services and the provider. There has also been a desire for an algorithm that can do this.

  Therefore, the present invention has been devised in view of the above-described problems. In addition to the evaluation from the user and the estimated amount of the provider, the positional relationship between the user and the provider is comprehensively considered. And an information posting system that can eliminate the extreme disparity according to the ranking that leads to the cancellation of the website. With the goal.

  The information posting system according to claim 1 is an information posting system for sequentially posting information on a provider who provides a product or service or an object to be provided on one screen of an information terminal, and performs past transactions with the provider. DR calculation means for calculating for each provider a rating rank R based on the evaluation value R obtained by quantifying the evaluation from the user and the advertising cost C of each provider, and the DR calculation for each provider. A probability distribution assigning means for assigning a probability distribution according to the posting rank DR calculated by the means, and determining the actual placement rank of each provider based on the probability distribution assigned to each provider, and based on the placement rank Display control means for sequentially displaying information on the provider or the provision target on one screen.

  The information posting system according to claim 2 is the invention according to claim 1, wherein the position information of the provider is received in advance and the position information of the user who desires to provide the product or service is acquired. Distance variable calculating means for calculating a distance variable L based on the information and the position information of the user, and the DR calculating means further includes the distance variable L calculated based on the distance variable calculating means. The publication rank DR is calculated for each provider.

  The information posting system according to claim 3 is the invention according to claim 1 or 2, wherein the DR calculation means further quantifies the cumulative posting period T posted on the information terminal and the past transaction performance of the provider. The publication rank DR is calculated for each of the providers based on any one or more of the number N of actual achievements and the contract variable P that represents the publication contract on the information terminal.

  The information posting system according to claim 4 is the invention according to any one of claims 1 to 3, in which the degree of similarity between the product or service provided by the provider and the product or service desired by the user is expressed numerically. A similarity quantification unit that converts the ranking into DR for each provider based on the similarity quantified based on the similarity quantification unit. Features.

  The information posting system according to claim 5 is the invention according to any one of claims 1 to 4, wherein the probability distribution assigning means assigns each provider to a base rank in descending order of the posting rank DR, Probability distributions set so as to be distributed based on the above-mentioned base ranks are respectively assigned.

  According to the present invention configured as described above, a provider with a high publication rank DR does not always have a higher actual publication rank, and there is a low probability that a provider with a lower publication rank DR has a lower rank. In some cases, the ranking is higher. For this reason, the provider whose publication rank DR is lower can solve the problem that a large disparity occurs between the provider whose publication rank DR is higher and the publication rank DR is lower. Registering with a service can be prevented from becoming meaningless in itself, and it is possible to prevent situations where the service is canceled one after another.

  In addition to evaluations from users who have made past transactions with the provider and the advertising expenses of the provider, the ranking is set comprehensively based on the positional relationship between the location information of the user who wants to provide products or services and the provider. It becomes possible.

1 is a system configuration diagram of an information posting system to which the present invention is applied. It is a figure for demonstrating the algorithm by the information printing system to which this invention is applied. It is a figure which shows the example which calculates an actual publication order and displays it in descending order on one screen. It is a figure which shows the example of the detailed screen required when determining whether to perform a transaction with a provider.

  Embodiments of an information posting system to which the present invention is applied will be described in detail below with reference to the drawings.

  FIG. 1 shows a system configuration diagram of an information posting system 1 to which the present invention is applied. The information posting system 1 includes a public communication network 2, a server 3, a provider terminal 4, and a user terminal 5 connected to the public communication network 2.

  The public communication network 2 is a communication network accessible from a personal computer (PC) or a portable terminal such as the Internet.

  The server 3 is a storage device for storing website tools (software) necessary for realizing the information posting system 1 to which the present invention is applied.

  The provider terminal 4 is an electronic device such as a PC or a portable terminal that is actually operated by a provider who desires to provide a product or service through a website. The provider here is not limited to a business that provides products and services as a business, but includes a business that provides these without business.

  The user terminals 5a and 5b are information terminals such as PCs and portable terminals that are actually operated by a user who desires to receive provision of products and services through a website.

  Such an information posting system 1 realizes the operation described below based on software stored in the server.

  First, the provider first registers information related to products and services provided to the user. The registration process may be performed on behalf of the system management side.

  As for the registration of this information, the advertising cost C related to the product or service to be provided and the location information of the provider are registered. The advertising cost C here is a service consideration paid by the provider to the management side. For example, in the case of an estimation site of a solar power generation system, it is an advertising fee charged to a supplier every time a quote request is transmitted from the user to the supplier. Further, in the case of a restaurant or cosmetic portal site, the concept may include an advertisement amount that is charged each time a user accesses a detailed page of each restaurant or cosmetic. Further, the advertising cost C may be a parameter indicating an estimated amount of money that the provider wishes to trade with the user under what financial conditions. And they are not limited to the case where they are embodied as monetary amounts. That is, as an alternative to the price of products and services provided by the provider, items that have values to the provider such as goods and services from the user may be included in the advertising cost C as an estimated amount. Further, the advertising cost C may include both such service consideration and an estimated amount. In such a case, these may be arbitrarily weighted and parameterized.

  Further, the location information of the provider is information regarding the provider's address and whereabouts, but is not limited thereto, and may be information regarding the current location. Further, the position information is not limited to the position information of the provider itself, and may be position information of a person acting on behalf of the provider.

  When the provider desires to register such advertising expenses C and position information together with products and services provided by the provider, the provider performs an input operation through the provider terminal 4. Then, the advertising cost C and position information input from the provider are registered in the server 3 via the public communication network 2.

  Incidentally, in the server 3, in addition to these pieces of information, information described below is registered for each provider.

  First, an evaluation value R obtained by quantifying the evaluation from a user who has made a past transaction with a provider is also registered. The evaluation value R may be embodied, for example, as a numerical value obtained by showing the evaluation of the provider by the user in a multi-step evaluation, or actually received a word-of-mouth evaluation from the user. If it is, it may be a numerical value. That is, the evaluation value R may be any value obtained by quantifying any evaluation by a user who has made a transaction with the provider.

  Further, location information of a user who desires to provide a product or service is also registered in the server 3. This user location information may be current location information, for example, for a user who operates the user terminal 5b, which is a mobile terminal, or for a user who operates the user terminal 5a, which is a desktop PC. It may be an address or residence.

  Thus, since the location information of the user and the location information of the provider are registered in the server 3, a distance variable L indicating how far the location of the user and the provider is separated is calculated. It is also possible.

  The server 3 is also registered with the cumulative posting period T for products and services provided by the provider. For example, the cumulative posting period T may be measured by registering the total posting time of the products and services provided by the provider, or a part of the total posting time may be weighted. The calculated value may be the cumulative posting period T. Further, the cumulative posting period T may be calculated for the cumulative posting time for each region.

  The server 3 may be registered with a record number N obtained by quantifying the past transaction record of the provider. The actual number N may be embodied as indicating how many times the provider has made a transaction with the user, or may be calculated by weighting a predetermined one of the transactions. Also good.

  The server 3 may be registered with a contract variable P that is a numerical value of a contract posted on the website. The contract variable P is information indicating under what conditions the provider allows his / her product or service to be posted on the website with the administrator who manages the information posting system 1. The contract variable P may indicate, for example, whether the provider has his / her product or service posted for a fee, or may be posted for free. It may be quantified according to the price plan. Furthermore, the value of the contract variable P may be reflected on other parameters. For example, in the case of a free fee plan, the cumulative posting period T may be calculated by about 1/5 of the posting period in the case of a paid plan.

  In this way, the server 3 has the advertising cost C, the evaluation value R, the distance variable L, the cumulative publication period T, the actual number N, and the contract variable P related to each provider, and the products and services provided by the provider. It will be registered together. In the server 3, first, among the various registered businesses, the server 3 performs grouping for each provider that provides the same kind of product or service. Incidentally, if this information posting system 1 accepts only the posting of the same kind of products or services, this grouping step will be omitted.

  Next, for this provider, the display rank DR (Display Rank) is set for each of the above-mentioned parameters (advertising cost C, evaluation value R, distance variable L, cumulative posting period T, actual number N, contract variable P). Calculate for the provider.

  Based on the calculated DR, the DRs of the respective providers are sorted in descending order to calculate the “basic rank” that is the base rank. Based on this base order, distributed display is performed based on a certain probability distribution.

  First, a method for calculating the posting rank DR will be described.

  When the posting rank DR is obtained for each provider, first, normalization processing is performed for each parameter (advertising cost C, evaluation value R, distance variable L, cumulative posting period T, actual number N, contract variable P). Normalization here means dropping these parameters into normalization variables in the range of 0 to 1. The reason for normalizing each parameter is that each parameter does not have the same unit, and therefore, in calculating the posting rank DR, there is a difference between affected parameters.

  As an example of normalization, for example, when advertising cost C is normalized, it may be approximated by a normal distribution function on the assumption that the population has a normal distribution. The normal distribution function is the advertising cost on the horizontal axis and the number on the vertical axis, but using the normal distribution function table, etc., the top 5% of advertising expenses are normalized variables = 1.00, the top 5-10% are normalized variables 0.95 and the top 10-15% are assigned with a 5% pitch so that the normalized variable = 0.90...

  As another example of normalization, when there are 100 registered providers, for example, they are sorted in descending order in the order of advertising expenses C. The column sorted in descending order is divided equally into 20 sections. When there are 100 providers, 5 companies are divided into 20 categories in descending order of advertising cost C. Next, normalization may be performed such that the first division from the top is normalized variable = 1.00, the second division is normalized variable = 0.95,.

  In this way, all parameters (advertisement cost C, evaluation value R, distance variable L, cumulative posting period T, actual number N, contract variable P) are converted into normalized variables in the range of 0-1. This conversion method is not limited to the above-described example, and any method may be used as long as it can be normalized to a range of 0 to 1. Hereinafter, the normalized advertising cost C is displayed as [C], the evaluation value R is normalized as [R], and the distance variable L is normalized as [L]. , The cumulative publication period T normalized is displayed as [T], and the actual number N is normalized as [N]. The normalized contract variable P is displayed as [P]. Here, it is assumed that [P] is set to 1 for a paid plan and 0 for a free plan.

  Next, the publication rank DR is calculated using this normalized variable. The following formula (1) is a formula for calculating the publication rank DR.

_{DR = (a 1 × [C} ] + a 2 × [N] + a 3 × [R] + a 4 × [L] + a 5 × [T] + [P]) × 5 ····· (1)

  Further, as a modification of the equation (1), the posting rank DR may be calculated based on the following equation (1) ′.

_{DR = (a 1 × [C} ] + a 2 × [N] + a 4 × [L] + a 5 × [T] + [P]) × a 3 × [R] ··· (1) '

Note that a ₁ + a ₂ + a ₃ + a ₄ + a ₅ = 1. As a result, the total value in parentheses in the formula (1) is in the range of 0-2. By multiplying this by 5, DR becomes a numerical value of 0-10.

[P] is not limited to the case of setting to 0 or 1 depending on whether or not it is the above-described paid plan, and may be normalized by dividing into three or more categories. In such a case, weighting may be performed by multiplying the coefficient a ₆ by this.

Also, for each coefficient of a ₁ ~a _5, the operator this information publishing system 1 may be set arbitrarily.

It should be noted that the above-described normalization is not particularly essential when obtaining the publication rank DR. That is, each parameter before normalization may be directly substituted in the above equation (1). At this time, by adjusting the weights of the coefficients a _{1 to} a ₅ , it is possible to realize the same effect as that of normalization.

  In addition, when determining the posting rank DR, it is limited to using all the parameters (advertising cost C, evaluation value R, distance variable L, cumulative posting period T, actual number N, contract variable P). It is not a thing.

  For example, the posting rank DR may be calculated for each provider based on the evaluation value R and the advertising cost C.

  In addition to the evaluation value R and the advertising cost C, the posting rank DR may be calculated for each provider based on the distance variable L.

  Furthermore, in addition to these three parameters, the posting rank DR may be calculated for each provider based on any one or more of the cumulative posting period T, the actual number N, and the contract variable P.

  After obtaining the posting rank DR for each provider in this way, the process proceeds to a process of determining the actual ranking of each provider based on the ranking algorithm described below.

  As one of the methods, first, as shown in Table 1, the posting rank DR in the range of 0 to 10 is divided into 20 sections from q = 1 to 20. Of course, the number of divisions is not limited to 20. However, as the number of divisions is increased, minute changes in DR are reflected in the display order. Normally, at most 20 sections are sufficient.

A subscript q is associated with each of the 20 divided DRs. The display order of companies belonging to the segmented DR follows the probability distribution f _q (x).

That is, this probability distribution f _q (x) is set in advance according to each q. Since this q corresponds to the posting rank DR as described above, in other words, the probability distribution f _q (x) corresponds to the posting rank DR.

As described above, since each provider is required to have a posting rank DR, a value of q corresponding to this posting rank DR is assigned. As a result, a probability distribution f _q (x) corresponding to each q is obtained. Assigned.

The probability distribution f _q (x) may be a normal distribution function, for example. The probability distribution f _q (x) is not limited to the normal distribution, and may be any probability distribution. The standard deviation σ of the normal distribution function represents the spread of the distribution, which is determined in advance from an empirical rule.

Next, for the publishers, the actual ranking is determined based on the probability distribution f _q (x) assigned to each as shown in FIG. The probability distribution f _q (x) is not limited to an upward convex shape, and may be any probability distribution. In FIG. 2, the vertical axis represents the actual placement order, and the horizontal axis represents the placement probability for each placement order based on the respective probability distributions f _q (x).

A publisher assigned with q = 1 has a very high probability that the actual ranking will be higher, but the ranking may be lower with a small probability. As q increases, the probability distribution f _q (x) gradually shifts to the lower order in the actual placement order. The publisher assigned with q = 20 has a very high probability that the actual ranking will be lower, but the ranking may be higher with a small probability.

  Thus, according to the present invention, a provider with a high ranking rank DR does not always have a higher actual ranking, and there is a low probability that a provider with a lower ranking rank DR also has a lower actual ranking. There are cases where it becomes higher.

  For this reason, the provider whose publication rank DR is lower can solve the problem that a large disparity occurs between the provider whose publication rank DR is higher and the publication rank DR is lower. Registering with a service can be prevented from becoming meaningless in itself, and it is possible to prevent situations where the service is canceled one after another.

  In addition to evaluations from users who have made past transactions with the provider and the advertising expenses of the provider, the ranking is set comprehensively based on the positional relationship between the location information of the user who wants to provide products or services and the provider. It becomes possible.

  Note that a user who actually desires to use the information posting system 1 accesses the website and inputs his / her position information and the like. As a result, the information posting system 1 calculates the actual ranking of the provider based on the above-described algorithm, and on the information terminal such as the user terminal 5 of itself, for example, one screen as shown in FIG. Above, information about the provider is displayed in descending order from the top. Naturally, the provider displayed from the top is more likely to be noticed by the user. When the user clicks a desired one of the displayed providers, the screen is switched to a detail screen as shown in FIG. Then, it is possible to determine whether or not to make a transaction with the provider through this detailed screen.

  By the way, after the transaction, the user can evaluate the provider, which becomes a so-called word-of-mouth and is reflected in the evaluation value R.

  The present invention is not limited to the embodiment described above. When determining the actual publication rank, the providers are sorted in descending order from the top based on the publication rank DR, and the base rank is assigned. For example, when there are 20 providers, base ranks 1 to 20 are assigned based on the calculated ranks DR. Then, a probability distribution set based on this base rank may be assigned. For example, when the base ranking is 1st, a probability distribution in which the probability that the actual ranking is higher is assigned as shown in FIG. 2 is assigned, and the actual ranking increases as the base ranking gradually decreases. Assign a probability distribution that gradually becomes lower.

  Moreover, in this invention, it is not limited to the case where the provider which provides a product or a service is listed in order on one screen on a website. Instead of listing the provider, the provision target (a product or service provided by the provider) may be posted in order on one screen on the website. The algorithm in such a case is the same as that described above, and the provision targets are displayed in descending order instead of the provider when actually placing.

  Further, the present invention is not limited to the case where the provider or the provision target is posted on a website. As an alternative to be posted on the website, it may operate on an application of a PC or a portable information terminal.

  According to the present invention, unlike the conventional ranking in which the advertising cost and the click rate are reflected, the word-of-mouth evaluation as a result of the user actually using the product or service such as a word of mouth can be reflected in the ranking. It becomes possible.

  In addition, according to the present invention, a display order combining word-of-mouth evaluation and location information is possible, and this is also useful in a store search service that uses the location information of the person using an information terminal. For example, the present invention is effective when a user uses a PC website for home renovation and wants to find a business that is close to the home and has good cost performance. In addition to the renovation, the present invention is applied in a wide range of fields such as searching for a photovoltaic power generation system sales company and a restaurant.

  As an example, assume a case where a renovation company is searched. In the conventional method, the user will search for stores by narrowing down “Kanagawa Prefecture> Yokohama City> Yokodai”. Then, the system side extracts a list of traders who have a location in the municipality or cover the municipality, and in the extracted list, for example, “In order of high word-of-mouth evaluation, advertisement It will be displayed in the form of "in order of expense".

  However, if the user narrows down to “Kanagawa Prefecture> Yokohama City> Yokodai”, for example, if only “Highest word-of-mouth evaluation” is displayed, vendors in Saitama Prefecture, Tokyo, and Chiba Prefecture are included. there is a possibility. This is because Saitama traders may also target Yokodai. However, from the user's point of view, it is presumed that a service provider such as renovation is more convenient for a service such as renovation and feels a sense of security psychologically.

  For this reason, in order to search for “a business that is close to home and has good cost performance”, in the conventional method, check the selection item of “a business in Yokohama city” or “ It is necessary to narrow down the condition of “traders within km”.

  However, in such a method, the narrower the distance, the higher the probability that a business with a high word-of-mouth evaluation will be excluded from the search results. It is necessary to sort (order) word-of-mouth evaluation. Moreover, it is inferior in convenience as a system that a user cannot find a target trader unless he / she thinks so far.

  In addition, even if the user is able to narrow the distance range and sort it in the order of word-of-mouth evaluation, there are many traders that are far from home because of the wide distance range. . For this reason, in order to find a vendor that is “close to home and has good reviews,” it is necessary to determine the location of each vendor from the top, making it cumbersome.

  On the other hand, according to the present invention, since the evaluation numerical value R and the distance variable L can be comprehensively taken into consideration, it is possible to determine and display the trader's ranking in a single shot. It is possible to provide a system with improved convenience for users who place importance on “word of mouth”.

  Further, according to the present invention, it is possible to display stores in a descending order with priorities, which are “close to the user's location, have high word-of-mouth evaluation, and have high advertising costs”. In other words, since it is possible to display in the order of evaluation of the middle of those instead of “advertising expenses” or “word-of-mouth evaluation”, the convenience for the user is greatly enhanced as compared with the conventional method. For example, when a plurality of narrowing conditions are set, it is often the case that “stores with such conditions do not apply” are displayed in the conventional method. However, in the present invention, since the above-described algorithm is adopted instead of narrowing down with “On or Off” for each condition, “corresponding zero” is not displayed, and the search condition set by the user It is possible to list more preferable stores along descending order. This increases convenience for the user and leads to an improvement in profit opportunities for the store.

  If “emphasis on distance” is selected, it can be handled by increasing the weighting of the distance numerical value. If it is “word-of-mouth emphasis”, by increasing the weight of the word-of-mouth evaluation, it is possible to display a very reputable store that is a little far away.

  If there is a condition that the user can't give up, the fuzzy function can be intentionally restricted by preparing items such as “This condition is required”.

  In addition, as described above, the present invention has a function of distributing the ranking using the probability distribution. For this reason, the concentration of popularity to stores with high word-of-mouth evaluation can be eased. For example, a website about food is easy to understand, but when you open a portal site of this kind, the first screen that appears is the “Total Ranking” screen. The display order of each store is fixedly displayed in descending order of the “total ranking”. In other words, no matter how many times the page is reloaded, the ranking is always fixed based on the total line king number, so only popular stores are always displayed at the top, and access is concentrated. There's a problem.

  For example, if a store has an evaluation score of 0 to 5.00 and there are 20 stores whose total score is between 4.00 and 4.20, basically, there will be no significant difference in the quality of any store. However, since only stores with a high score are always displayed at the top, even though the store with a score of 4.00 is also high in quality, this store is ranked 20th, so there is a drawback that there is almost no access from the user.

  In this way, if the ranking is fixed based on only one score, as in the conventional method, especially if there is a limit on the number of people such as stores, extreme concentration occurs and congestion occurs. In addition, convenience for the user is impaired. At the same time, it is also a disadvantage that stores that are inferior in terms of scores but are really good are not accessed in a much lower rank. On the other hand, according to the present invention, it is possible to avoid the disadvantage of “access concentration”. In particular, distributed display is an important function for sites based on word-of-mouth evaluation. Specifically, by calculating the base rank by combining various parameters and displaying the base rank as a reference with a probability distribution such as a normal distribution, for example, the problem of extreme concentration can be avoided. Note that the normal function is used as the probability distribution because the normal function is well-known. In principle, any function can be used as long as the curve is convex. Display is feasible.

1 Information posting system 2 Public communication network 3 Server 4 Provider terminal 5 User terminal

Claims (5)

In an information posting system for sequentially posting information on a provider providing a product or service or an object to be provided on one screen of an information terminal,
DR calculating means for calculating for each provider a rating rank R based on an evaluation value R obtained by quantifying the evaluation from a user who has made a past transaction with the provider and an advertising cost C of each provider;
Probability distribution assigning means for assigning probability distributions according to the posting rank DR calculated by the DR calculating means to the providers,
Display control means for determining the actual ranking of each provider based on the probability distribution assigned to each provider, and sequentially displaying information on the provider or the provision target based on the ranking. An information publishing system characterized by comprising The distance for receiving the location information of the provider in advance, acquiring the location information of the user who desires to provide the product or service, and calculating the distance variable L based on the location information of the provider and the location information of the user A variable calculation means;
The information posting system according to claim 1, wherein the DR calculating means further calculates the posting rank DR for each provider based on the distance variable L calculated based on the distance variable calculating means. The DR calculating means further includes a cumulative posting period T posted on the information terminal, a number N of results obtained by quantifying the past transaction results of the provider, and a contract variable P that represents the number of posted contracts on the information terminal. The information posting system according to claim 1, wherein the posting rank DR is calculated for each of the providers based on any one or more. It further comprises similarity degree quantification means for quantifying the degree of similarity between the product or service provided by the provider and the product or service desired by the user,
The DR calculation unit calculates the posting rank DR for each provider based on the similarity quantified based on the similarity quantification unit. Information posting system according to item 1. The probability distribution assigning means assigns each provider to a base rank in descending order of the posting rank DR, and assigns each probability distribution set to be distributed based on the base rank. The information posting system according to any one of?

Images