The essay subsystem 105 and the essay editing subsystem 110 are discussed in greater detail in FIG. 2 and FIG. 3, respectively. The essay prompts 104 that the essay sorting subsystem 105 sorts may comprise essay prompts for applying to the personalized school list 102. The application subsystem 103 may generate a plurality of essay prompts taken from the schools in the list. For example, if UC Berkeley is a school on the personalized list of schools, the personalized list subsystem may go to the essay prompts that UC Berkeley asks for in an application and add the prompts to the list of prompts the essay sorting subsystem will be applied to. The essay sorting subsystem 105 may use essay prompts input by the user 111. The essay sorting subsystem 105 may comprise a natural language processing (NLP) module 106 and a clustering module 107, as described herein. The NLP module 106 may implement a natural language processing algorithm that may process the essay prompts to generate features. The clustering module 107 may implement a clustering algorithm that processes the features to categorize the essay prompts into one or more subsets.

The subsystems of FIG. 1 and their components can be implemented on one or more computing devices. The computing devices can be servers, desktop or laptop computers, electronic tablets, mobile devices, or the like. The computing devices can be located in one or more locations. The computing devices can have general-purpose processors, graphics processing units (GPU), application-specific integrated circuits (ASIC), field-programmable gate-arrays (FPGA), or the like. The computing devices can additionally have memory, e.g., dynamic or static random-access memory, read-only memory, flash memory, hard drives, or the like. The memory can be configured to store instructions that, upon execution, cause the computing devices to implement the functionality of the subsystems. The computing devices can additionally have network communication devices. The network communication devices can enable the computing devices to communicate with each other and with any number of user devices, over a network. The network can be a wired or wireless network. For example, the network can be a fiber optic network, Ethernet® network, a satellite network, a cellular network, a Wi-Fi® network, a Bluetooth® network, or the like. In other implementations, the computing devices can be several distributed computing devices that are accessible through the Internet. Such computing devices may be considered cloud computing devices.

FIG. 2 is a flowchart of an example process for generating groups of similar essays. The process can be performed by the essay sorting subsystem 105 of FIG. 1, which may be implemented on one or more computers in one or more locations.

The essay sorting subsystem can obtain a plurality of essay prompts (210). The plurality of essay prompts may comprise textual essay prompts (e.g., a text document, a portion of a webpage), essay prompts from scanned documents (e.g., documents that were mailed to a user), or other prompt sources (e.g., verbal transcription). The scanned documents may be transformed into textual essay prompts by an optical character recognition (OCR) process. The essay sorting subsystem may obtain an essay prompt of the plurality of essay prompts by scrapping the prompt from a website (e.g., an algorithm that downloads data from the website of a college), retrieving the essay prompt from a database, retrieving the essay prompt from a single application service (e.g., the common application), or accepting the prompt from the user (e.g., a user submitted prompt). Each essay prompt of the plurality of essay prompts may be obtained by any method (e.g., all essay prompts of the plurality of essay prompts do not need to be obtained the same way).

The essay sorting subsystem may extract one or more features from each essay prompt using a natural language processing (NLP) algorithm (220). The one or more features may comprise at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 50, 100 or more features. The one or more features feature may comprise the subject of the essay prompt, the call of the essay prompt, the category of the essay prompt, the topic of the essay prompt, the allowed length of the essay, or any combination thereof. In some cases, the feature may be a semantic feature. The features may be quantitative features. The NLP algorithm may be a rule-based algorithm, a statistical algorithm, a machine learning algorithm, a semantic analysis algorithm, or any combination thereof. For example, the essay prompt can have a machine learning algorithm applied that determines not only what the question said literally, but also the intended call of the question.

Using the one or more features, the essay sorting subsystem can generate two or more subsets of the plurality of essays prompts (230). The two or more subsets may comprise at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 25, 50, or more subsets, up to the number of essay prompts 210. The essay sorting subsystem may minimize the quantity of the two or more subsets. The user may instruct the subsystem about a maximum number of subsets. An essay prompt in a subset may satisfy a measure of similarity to other essay prompts in the subset. The measure of similarity may comprise all essays having a value of at least one feature that is above a threshold value, having a value of at least one feature that is substantially similar, or a combination thereof. The measure of similarity may be a quantitative measure.

The essay sorting subsystem may generate the subsets by applying a clustering algorithm to the one or more features. The clustering algorithm can comprise a hierarchical clustering algorithm. A hierarchical clustering algorithm may be a clustering algorithm that clusters objects based on their proximity to other objects. For example, a hierarchical clustering algorithm can cluster essay prompts based on the proximity of NLP features from one essay prompt to corresponding NLP features from other essay prompts. The clustering algorithm can alternatively be a centroid-based clustering algorithm, e.g., a k-means clustering algorithm. A k-means clustering algorithm can partition n observations into k clusters, where each observation belongs to the cluster with the nearest mean. The mean can serve as a prototype for the cluster. In the context of NLP features from the subsystem 200, a k-means clustering algorithm can generate distinct groups of essay prompts that have NLP features that are correlated. The clustering algorithm can alternatively comprise a distribution-based clustering algorithm, e.g., a Gaussian mixture model or expectation maximization algorithm. Examples of other clustering algorithms that the subsystem 200 can train and implement are cosine similarity algorithms, topological data analysis algorithms, and hierarchical density-based clustering of applications with noise (HDB-SCAN).

The essay sorting subsystem may make recommendations as to the essays to be written for the two or more subsets. The recommendations may comprise a topic of the essay or a theme to be included in the essay. The essay sorting subsystem may generate a report on the applicability of at least one of a user's written essays to at least one of the essay prompts. For example, a student wrote 5 different application essays for 5 different prompts and then wants to have an essay for a sixth prompt. In this example, the essay sorting subsystem analyzes the new prompt and recommends one of the already written essays to answer the new prompt.

FIG. 3 is a flowchart of an example process for editing an essay. The process can be performed by the essay editing subsystem 110 of FIG. 1, which may be implemented on one or more computers in one or more locations.

The essay editing system can receive one more documents from a user (310). The one or more documents may be college admissions essays, personal statements, resumes, writing samples for an application, or financial aid application essays. The documents may be one or more types of documents. For example, a user can submit both a personal statement and an admissions essay to the marketplace.

The essay editing subsystem can associate the one or more documents with an editor (320). In some cases, the essay editing subsystem may associate the one or more documents with an editor automatically. For example, the essay editing subsystem may select one of a plurality of designated resume editors to edit a resume. The selected resume editor may meet one or more criteria, e.g., an availability criterion, a rating criterion, a price criterion, a language proficiency criterion, a subject expertise criterion, if the editor has worked with the user before, or the like. In other cases, the essay editing subsystem may associate the one or more documents with an editor in response to an editor selecting or claiming the essay on the college admissions platform.

The editor claiming the documents may be one of a plurality of editors for a plurality of documents. For example, one editor may claim a user's admissions essay while another editor may claim a user's financial aid essay. Alternatively, one editor may claim a plurality of documents. The editor may be an expert in the subject matter of the essay. The editor may have a limited time to complete the editing of the one or more documents. The editor may claim the one or more documents through a user interface, such as, for example, a graphical user interface (GUI).

The association of the editor with a document may prevent other editors from claiming the document. The association may be for a limited time. The association may be revoked by the user, the editor, or an administrator of the marketplace. The association may last for one or more editing cycles. For example, a user may like the work a particular editor is doing, so the user can choose to have the same editor perform an additional round of edits after the first.

The essay editing subsystem can receive edited versions of the one or more documents from the editor (330). The edited version may comprise the original one or more documents with one or more added deletions, insertions, comments, other feedback, or any combination thereof. The edited versions may be generated according to instruction from the user.

The essay editing subsystem can then display the edited versions of the document to the user (340). The documents may be displayed to the user in a GUI. The user may download the documents. The user may be able to comment on the quality of the edited documents. The user may be able to interact with the editor to clarify the edits to the documents. The user may be able to leave a review of the editor, which may comprise a sore and/or a written review.

The edited document may be further analyzed by an algorithm. The algorithm may comprise an algorithm as described herein, such as, for example, a natural language processing algorithm. The edited versions of the one or more documents may comprise emotional tone scores that may indicate a likelihood that a reader of said one or more documents may experience one or more emotions. The emotional tone score may be generated by the algorithm. For example, the algorithm is applied to an edited document, and it produces a score of the likelihood that the document will make a reader feel joy. The algorithm may generate an emotional tone score on both the edited and unedited versions of the one or more documents. The user may use the emotional tone score as at least a part of a determination if a desired emotion is conveyed by the one or more documents.

FIG. 4 shows an example user interface 400 for the essay editing marketplace. The user interface may enable a user to initiate the creation of a new document to be placed on the essay editing marketplace, for example with a button 410. The user interacting with the button may open a different user interface. The user may be able to keep track of the essays that have been started or submitted using a list 420. The list may have at least one piece of information, such as the user generated name of the essay, the type of the essay, the date of submission, the status of the essay edits, the due date of the essay, and the school the essay is for. The list may be ordered with respect to any one of the pieces of information. For example, the list may be ordered by the due date of the essay where the nearest due date appears first. The list may further comprise a rating 430 given by the user to the quality of the edits made by the editor. The rating may comprise a numerical rating, a graphical-based rating (e.g., stars), a written rating (e.g., a review), or any combination thereof. The user may be apprised of the details of how much editing the user has paid for in another part of the interface 440. The user may subscribe to the essay editing marketplace (e.g., pay by the month), pay by the word for each essay submitted, pay by the number of essays submitted, or any combination thereof.

The essay editing subsystem may comprise a computer implemented marketplace. The marketplace may comprise an online marketplace. The marketplace may have a standard cost to a user for an editor to edit the user's one or more documents. For example, a personal statement can cost the user $300 to be edited. The marketplace may have a scaling cost to the user for the edits. For example, a personal statement can cost $50 per page to be edited. The marketplace may have different costs to the user depending on the editor who claims the user's one or more documents. For example, three different editors can have rates of $100, $150, and $200 per document edited, respectively. The user may set a maximum budget, a minimum budget, or both for the total cost of the editor editing the user's one or more documents. For example, a user may say that they are willing to pay no more than $100 for their financial aid application to be edited. In this example, a number of editors may choose not to claim this document, as the payment may be too low.

FIG. 5 shows an example user interface 500 for submitting an essay to the essay editing marketplace. A user may access the interface 500 by interacting with a different element of another user interface (e.g., pressing a button). The interface may comprise information 510. The information 510 may comprise instruction on using the essay submission interface, on writing an essay, on best practices for writing an essay of quality, or the like. The interface may display information about the user's essay 520. The information 520 may comprise word count, page length, estimated cost of editing, compliance with the formal requirements of one or more schools, or any combination thereof. The interface may comprise an area to prepare an essay for submission 530. The area 530 may comprise a space for a title, the relevant essay prompt, the desired word count, additional comments, requests, or any combination thereof. The essay may be uploaded (e.g., a document on the user's device may be transmitted to the marketplace) or written within an element of the area. The area may comprise an element to submit the essay, such as, for example, a button.

FIG. 6 shows an example emotional tone score display 600. The emotional tone score may comprise a representation of the emotional tone score 610. The representation 610 may comprise a graphical representation, a textual representation, a numerical representation, or any combination thereof. The representation may comprise at least one tone. For example, the representation may show the likelihood a reader of an essay would experience fear or joy. The representation may display any number of tones up to the total number of tones detected in the essay. The representation may have a threshold value that is exceeded to display a tone. The display 600 may further comprise an explanation of the emotional tone score 620. The explanation 620 may comprise a numerical score that may indicate the likelihood a given tone may be perceived by a reader. The explanation 620 may further comprise an additional explanation of what the tones scored in 610 mean. For example, the explanation can explain to the user what fear and joy mean.

FIG. 7 shows an example of an interface for a research request marketplace 700. The interface 700 may be accessed by a user. The interface 700 may allow a user to initiate the generation of a new research request via an element of the interface, such as, for example a button 710. The user interacting with the method to initiate creation of a new research request may open a different user interface. The user may be able to keep track of the research requests that have been started or submitted using a list 720. The list may comprise at least one piece of information, such as the user generated name of the research request, the type of research request, the date of submission, the status of the research request, the due date of the research request, and the topic the research request is for. The list may be ordered with respect to any one of the pieces of information. The list may further comprise a rating 730 given by the user to the quality of the answer given to the user. The rating may comprise a numerical rating, a graphical based rating (e.g., stars), a written rating (e.g., a review), or any combination thereof.

FIG. 8 shows an example user interface for submitting a research request 800. The user may give a title to the research request 810. The title 810 may be mandatory. The user may provide information on the research request 820. The information may comprise the research request, information about the user, a due date of the research request, information about the topic of the research request, or any combination thereof. The user may submit a research request using an element of the interface 800, such as, for example, a button 830. The user may see the number of research requests the user has paid for in another element 840. The user may pay a subscription, pay by the research request submitted, or a combination thereof. The user may pay more for an urgent research request.

FIG. 9 shows an example user interface for a personalized college list 900. The interface 900 may comprise a collection of colleges 910. The collection 910 may further comprise information about the colleges of the collection of colleges. The information may comprise application due dates, applicant assigned qualifiers (e.g., if a college is a “safety school”), applicant provided notes, or any combination thereof. The collection 910 may be sorted and/or filtered by any one of the information categories. Additional colleges may be added to the collection 910 via an additional element of the interface, such as, for example, an entry line 920. The entry line 920 may comprise a plurality of categories for searching new colleges, including college name, location, attributes (e.g., selective, private), or any combination thereof. The entry line 920 may comprise a selectable deadline (e.g., regular decision, early decision) and/or an applicant assignable qualifier (e.g., if a college is a “reach school”). The entry line 920 may further comprise an element to add the new college to the collection, such as, for example, a button. The interface 900 may further comprise a representation of the collection 930. The representation 930 may be graphical. The representation may be dynamic. The representation may be of the applicant assignable qualifiers of the collection 910 (e.g., how many schools are “reach schools” versus “safety schools” versus “target schools”). The representation 930 may comprise other information, such as the distribution of deadlines, of school costs, of school location, of undergraduate populations, or the like. The representation 930 may be interactive. For example, a user can click on the portion labeled “high reach schools” and the collection 910 would be filtered to show only “high reach schools.”

FIG. 10 shows an example detail view of a college 1000. The view 1000 may be accessed by interacting with an element of FIG. 9 collection 910. The view 1000 may comprise a summary 1010. The summary may comprise information about the school (e.g., contact information, location, university type) and/or a brief summary of school statistics. The view 1000 may further comprise detailed information 1020. The detailed information may comprise admission statistics, enrollment statistics, academic statistics, campus statistics, research statistics, faculty statistics, any combination thereof, and the like. The detailed information may further comprise additional information accessible via an element of the interface, such as, for example, a button 1030. The additional information may provide further detail (e.g., admitted applicant GPA, school sports teams).

Computer Systems

The present disclosure provides computer systems that are programmed to implement methods of the disclosure. FIG. 11 shows a computer system 1101 that is programmed or otherwise configured to implement the system 100 for FIG. 1. The computer system 1101 can regulate various aspects of the present disclosure, such as, for example, the process 200 of FIG. 2 and/or the process 300 of FIG. 3. The computer system 1101 can be an electronic device of a user or a computer system that is remotely located with respect to the electronic device. The electronic device can be a mobile electronic device.

The computer system 1101 includes a central processing unit (CPU, also “processor” and “computer processor” herein) 1105, which can be a single core or multi core processor, or a plurality of processors for parallel processing. The computer system 1101 also includes memory or memory location 1110 (e.g., random-access memory, read-only memory, flash memory), electronic storage unit 1115 (e.g., hard disk), communication interface 1120 (e.g., network adapter) for communicating with one or more other systems, and peripheral devices 1125, such as cache, other memory, data storage and/or electronic display adapters. The memory 1110, storage unit 1115, interface 1120 and peripheral devices 1125 are in communication with the CPU